|
 Stock Forums
 04. Best-Charts Software
 To Alvin (and possibly others) - BC and TA notes (Page 1)
|
 UBBFriend: Email This Page to Someone! This topic is 2 pages long: 1 2 |
next newest topic | next oldest topic |
| Author | Topic: To Alvin (and possibly others) - BC and TA notes |
|
david_louisson Member Posts: 303 |
posted 08-17-2004 10:58 PM
 
BEST-CHARTS AND TECHNICAL ANALYSIS PRIMER ===================================== DISCLAIMER Please note that I am in no way involved in the development of, nor do I have a financial stake in, Best-Charts. Any knowledge that I have of Best Charts has been gained simply by repeatedly trialling its different functions, and the 4 years that I have been studying trading and technical analysis. All of my comment here is submitted free of charge, on a take-it-or-leave-it basis. I accept no responsibility for trading decisions the reader might make in attempting to apply any of these principles or ideas. I seek no reward for any gains that readers might make, and likewise accept no liability for their losses. I do believe that, as a technical analysis tool, Best Charts offers excellent value for money, and I wish everybody MASSIVE success with their trading!! David Louisson
[NOTE: Whenever new material is added, its content will be listed (in reverse date sequence) here. I will also add a dummy post to the bottom of this thread, to ensure that the thread remains current.] 13 Dec 2006 - 17 Nov 2006 - 14 Aug 2006 - 09 Apr 2006 - 16 Mar 2006 - 09 Mar 2006 - 06 Mar 2006 - 08 Feb 2006 - 22 Jan 2006 - 15 Jan 2006 - 04 Jan 2006 - 01 Jan 2006 - 18 Dec 2005 - 17 Dec 2005 - 04 Dec 2005 - 02 Dec 2005 - 30 Nov 2005 - 10 Jul 2005 - 18 Jun 2005 - 12 Jun 2005 - 08 Mar 2005 - 01 Mar 2005 - 25 Feb 2005 - 23 Feb 2005 - 19 Feb 2005 - 17 Feb 2005 - 13 Feb 2005 -
30 Jan 2005 - 16 Jan 2005 - 15 Jan 2005 - 13 Jan 2005 - 12 Jan 2005 - 11 Jan 2005 - 4 Jan 2005 - 30 Dec 2004 - 19 Dec 2004 - 17 Dec 2004 -
This thread now includes the following essays on how to use Best-Charts (BC), along with some Technical Analysis (TA) tips and approaches. There is also a vast number of links to some of the best FREE instructional material on TA and trading, that is available on the Internet. (See the links in Essays # 1.6, 6.1, 6.2, 8.2, 11.1)
1.1 – Getting Started with Best Charts 2.1 – Changes in Version 4.33.2 3.1 – Using Historical Quote Files 4.1 – Installing New Versions of Best Charts 5.1 – Tips for a Struggling Trader 6.1 – Free Material on Trading Psychology 7.1 – Interpreting the Output Generated by Best-Charts 8.1 – Importance of Stop Losses and Position Sizing 9.1 – Multi-AIO 10.1 – How Critical is Correct Parameter Calibration in Terms of Overall Success? 11.1 – Understand The Indicators Used in Best Charts 12.1 – Navigating Yahoo Finance 13.1 - Does Technical Analysis really work? 14.1 - Index of Alternative Data Suppliers 15.1 - Changes in Best-Charts version 4.40 16.1 - Best-Charts menu and toolbar options
INDEX OF ESSAYS SORTED BY TOPIC: BEST CHARTS OPERATION & DISCUSSION INDICATORS VERSION CHANGE NOTES TRADING MISCELLANEOUS * these Essays contain useful links to a raft of relevant material available on the Internet ========================================= Alvin The following includes material from other posts of mine on this forum that you might find helpful. Some of these are old but the many of the principles should still apply. I haven’t edited these, so some of the material may be repeated in more than one of the posts. Good luck David Louisson
BC is an extremely powerful analysis tool and it is down to each trader to experiment to find the settings that deliver the best results, depending on the stocks, instruments, time frames, and market conditions that one is trading. What follows is a suggestion to get you started – you will need to develop it further from here on your own. Also, please note that BC is not my primary TA system, and that I haven’t tested what follows here, in situations where real money is being staked. First you must decide upon the time frame you want to trade. In my view (assuming you want to use daily rather than weekly price bars) BC works best for ‘swing’ type trading, i.e. your average time in a position could be between 5 to 20 days, depending on the market’s cycle lengths. As a starting point, click the BT button and set the number of quotes (price bars) to 88. This value was recommended somewhere else (I can’t remember where), but it allows for back-test/optimization across 4 months data @ 22 market days per month. Later you can experiment with this value to suit yourself. If you set this too high, any optimization will be giving the oldest data in the window equal priority with the most recent. If you set this too low, then you will not be analyzing enough price bars to attain statistical significance. Buy/sell price – I am looking to open positions at the start of each day, so the most true-to-life option for me is ‘opening price the following day of the signal’. There is nothing wrong with the other 2 options, but the ‘average price on signal day’ probably aligns itself best with intraday trading. Long/short – are you going to be trading long positions (i.e. exploiting rising trends), short positions (falling trends), or both? Indicators – doesn’t matter what this is set to. At this point, we are simply setting the other parameters for back-test/optimize (AIO). Click OK and close any HTML page that is generated. Now you have to decide on if, and how, you will optimize. Load a stock for TA, and then click the AIO button. I would start by setting all (including BBI) indicator check boxes on. Click OK and wait for the optimize to run. Then re-click the BT button and click OK. The ‘Gain% / Trade’ column in the second table displayed will give you an idea which indicators are perhaps most likely to give the best results for the stock being analyzed. If the ‘Number of Trades’ value is high, then the usefulness of the value in the ‘Total Gain%’ column is undermined, because BC does not take the (frequent number of) transaction costs into account. In any case, it is important to realize that, because the optimization is performed ‘in hindsight’ that the Gain values are ‘maximized’ rather than what could be realistically achieved in trading. This is because their real purpose is to provide an objective basis from which indicators, and different parameter settings, can be compared against each other. If an indicator is generating a huge number of (e.g. 40 or more) buy/sell signals across the 88 day period, then it may work better unoptimized. Click AIO to turn optimization off, and re-optimize. This indicator will be plotted according to its ‘factory’ settings (Parameter Set 1: see Stocks/Change Parameters of Indicators). You can also experiment to set your own custom parameter settings for each stock, and each indicator, in a file called opd.txt – see Essay # 1.4 below for details. Where you take optimization from here is your decision. You might decide to run AIO across several different stocks, noting which indicators give greatest profit across the whole group, or you might choose to use different indicators for each stock. You can change the weighting that each indicator contributes to BBI (menu option: Stocks/Change Weights for BBI) accordingly, and then use BBI (which is always a summary of all of the other indicators) as the basis of when to buy/sell (see some ideas below). If you find that an indicator is giving poor signals, and you want to exclude it from consideration, give it a weighting of zero. See Essay # 1.3 below for more details on the composition of BBI. Now some ideas about how to use BBI. You may decide to (1) wait for BBI to generate buy and sell signals, (2) wait for bullish/bearish count or advance/decline probability to reach a certain value, or (3) for an earlier, more aggressive entry trade when the BBI starts to move out of overbought/oversold territory (i.e. from the upper of lower extreme of the its chart). Greater probability of success is likely when BBI is moving from an extreme area, and even more so if the indicator has been ‘floating’ in the extreme area for an extended duration, as this indicates a higher degree of ‘oversoldness’ or ‘overboughtness’. It is important to realize that there is no ‘Holy Grail’ set of parameters that work well for every stock – you need to experiment to discover which indicators, and parameters give optimum profits for each stock, given current market conditions, and your style of trading. It is a case of understanding the facilities available in BC and what they mean, how to effect them via the menu system, and how to interpret the results. From there, it is a trial-and-error process of ‘curve fitting’, by optimization and weighting, to maximize potential profit.
BC’s portfolio facility can be very useful for displaying the ‘background climate’ against which your prospective trade will be made. I have set up a portfolio with the following international indices – Then I run ‘Analyze 20 stocks (quotes from web)’ and load the resulting HTML directly into Excel. I type formulae to sum the following 6 columns in the top table: ‘Probability Advance/decline’, ‘Signal number: Buy/sell’, ‘Trend: Bullish/Bearish’. Comparing these numbers on a day-to-day basis gives a feel for the world market direction. Also, ‘Signal number’ is the change from yesterday to today; ‘Trend’ is the total number of Bullish and Bearish, i.e. whether the last signal was a buy or a sell; ‘Candle’ lists any significant candle patterns that have occurred today. This portfolio can also help with intraday trading, by opening the above portfolio, and running ‘Intraday MicroCharts’. For example, I trade UK which is 5 hours ahead of US. The FTSE generally runs in harmony with the Nasdaq while both markets are simultaneously open. But then the final 5 hours trading in the US can give an indication as to what the FTSE might do the following day. Likewise, I look at the Asian markets, as what has happened during the day (I live in New Zealand) can also impact on the FTSE. I also set up portfolios for different UK sectors, and run the same kind of analysis. So that I can see which sectors are out-performing others, and where the UK market as a whole is heading. You can exploit the fact that certain markets frequently move in harmony with each other, and increase the probability of success by trading when more of the following elements are in agreement: Obviously keeping an ear on political and economic news can also provide clues, although I like to see confirmation in price movements before acting upon them.
Best Charts (BC) gives these two data items: 1) Buy Sell signals: these are the number of green Os (buy signals) and red Xs (sell signals) that have occurred TODAY, summed across all of the 15 charts. Across the top of each chart is the same text ‘Number of Signals: Buy …Sell’. This will contain a 1 against either Buy or Sell if the relevant signal has occurred TODAY. These are the component values that are being summed. Note that the buy and sell signals for each chart are usually generated when the relevant indicator crosses a moving average of itself. You can use the optimization feature (AIO button on the toolbar) to calibrate these indicators’ parameters to the values that would have given the highest profit over a historical back-test. That is the essence and the power of BC. Which parameter set was used to generate the charts is shown in lavender coloured font 'Parameters: '. Set 1/2/3/4 or AIO if optimized. (Note – see Essays # 1.4, 1.6 below for more information). Of the component indicators, many of these are the most popular offered by mainstream charting packages, but the developer of BC has included two of his own invention – ‘Intelligent Technical Analysis’ (ITA) and ‘Best Charts Indicator’ (BCI). These are proprietary – only the developer knows their underlying formula. The full outline of how the buy/sell signals for each chart are generated is as follows – 2) When a buy signal occurs, the chart is considered BULLISH until a sell signal occurs. Then it becomes BEARISH until the next buy signal occurs. You can see this at the right of each chart, where it says ‘Trend: Bullish’ or ‘Trend: Bearish’. The ‘Trend Summary’ at the top of the page is the sum of the number of charts that are Bullish or Bearish, and the Advance/Decline Forecast probability is calculated from this value (although I’m not sure exactly how). The greater the number of bullish charts, the higher the probability of advance, and vice versa. The topmost chart is the Bullish/Bearish Indicator (BBI). This is a summary of the current Bullish/Bearish values of all of the other charts, and it generates its own buy/sell signals when the two exponential moving averages (EMA) cross each other. By default, each indicator makes the same contribution (weighting factor = 1) to BBI, but you can change this (menu option: “Stocks/Change weights for BBI”. By running a back-test (BT button on the toolbar), you can see which indicator(s) would have returned the highest profit across the back-test period, and adjust the weightings if you wish. All of this flexibility means that how you use these indicators to determine when to trade is entirely your decision. It depends largely on the time frame you are trading (e.g. intraday, swing, longer term). You may decide to wait for BBI to generate buy and sell signals, wait for bullish/bearish count or advance/decline probability to reach a certain value, or (for an earlier, more aggressive entry) trade when the BBI starts to move out of overbought/oversold territory (i.e. from the upper or lower extreme of the its chart). INDICATOR TYPES USED BY BC a) Trend following – MA, MACD, DMI b) Momentum oscillator – SO, RSI, ROC, CCI, WMS, PPO c) Volume based – EFI, MFI, OBV, PVT d) Proprietary – ITA, BCI e) BBI – summary of all of the above
The parameter settings set the calibration for the indicators used by BC. As an example of what I mean by calibration, a 10-day RSI and a 14-day RSI are different calibrations of the RSI indicator. These would be expressed as RSI(10) and RSI(14) respectively. In BC, each indicator generally uses the crossover points of a calibrated indicator, and a calibrated exponential moving average (EMA) of that indicator, as the buy signals (green O’s) and sell signals (red X’s) for the stock. Altering the calibrations (or ‘parameters’) changes the shape of the two curves, causing the crossovers, and hence the signals, to occur at different points. BC allows you to use either fixed or optimized calibrations for each indicator. FIXED CALIBRATION OPTIMIZED CALIBRATION NOTE 1 – How to set up a FIXED calibration that applies to all stocks NOTE 2 – How to set up OPTIMIZED calibration 1. Load the stock’s chart by typing its ticker code in the ‘Symbol’ window, selecting a country from the ‘Site’ dropdown, and then clicking either ‘W’(Read historical quotes from website) toolbar button or the ‘H’ (read historical quotes from ASCII file) button, depending on your source of data (Yahoo website or ASCII text file). 2. Click the ‘BT’ (Back testing) button to set up the defaults for back-testing optimization. Enter the number of quotes (i.e. most recent price bars) you want BC to use in the back-test. You can have profitability based on entries at the average price, closing price or tomorrow’s opening price – select accordingly. Finally select whether you want either long, short or both types of trade used in the profitability calculation. In this instance, which indicator you check in the lower part of the dialog is irrelevant, since we are setting up an optimization. Click OK, and ignore the results of the back test. Return to BC. 3. Click the ‘AIO’ (All Indicator optimization) toolbar button to run the optimization. Check each of the indicators you want to optimize. Click OK and watch the progress bar as the optimization proceeds. This can take anything from 10–40 seconds. When finished, the dialog (magenta typeface) alongside the BBI chart should show ‘Parameters: AIO’, and all of the charts will have been re-drawn with the optimized Buy/Sell signals. You can see how each indicator has been optimally calibrated by looking at the text dialog just above its chart, e.g. RSI(14) RSI MA(10). 4. After you have run the above steps, you can OPTIONALLY save the optimized values for this stock, for future use, via the menu option “Stocks/All Indicator Optimization/Save Optimal parameters” (choose file name ‘opd.txt for daily quotes’ and click OK. If you do this, every time you load the stock’s chart (see step 1), these optimized calibrations will be loaded automatically and the dialog (magenta typeface) alongside the BBI chart should show ‘Parameters: opd.txt’. opd.txt is the name of the file (in the folder \M-C\) where these calibrations are permanently saved. There is one entry in opd.txt per stock, so that you can save different optimized settings for each stock. Note that opd.txt overrides all other parameter settings for a stock. Once an entry has been made for a stock in opd.txt, these parameters will always be used when its chart is loaded. To have that stock return to using parameter set 1, 2, 3 or 4 you must delete its entry from opd.txt, either by using a text editor or via menu option “Stocks/All Indicator Optimization/Delete Optimal parameters” (type in the stock’s ticker, select opd.txt and click OK). Note also that the opd.txt file should be used for daily price bar analysis; other files (opw.txt etc) can be used for different timeframes (weekly or intraday bars). Finally, note that the values entered in steps 2 and 3 above can not be saved in the same way that the indicator parameters are. Their latest values remain active for the session, but have to be re-entered every time BC is exited and re-loaded. NOTE 3 – How to permanently save different FIXED calibrations for each stock NOTE 4 – The param432(1,2,3,4).txt and opd.txt files That’s it folks! All of the above was discovered by my own trial-and-error process. Rik or Admin1, if any of this is incorrect, please post a correction. Also, please feel welcome to copy anything useful into any future user manual.
I have re-read hexachord’s initial post, and its ’11 steps’ are more of a BC operations overview, than a description of TA principles and techniques, let alone how to apply them to trading. Of course we would all like to be given a set of parameters that would consistently deliver a net gain over a prolonged period. However, my experience (about 4 years) with the markets (specifically, the UK market) suggests that this is an unrealistic expectation. There is no ‘Holy Grail’ system. Even the most successful traders encounter periods where their losses for the period outweigh their wins. Likewise, there is no single optimum set of parameters. You will need to experiment with different indicators and parameters to find those which best suit your chosen stocks, markets, time-frames and instruments (stocks, options, futures, etc). That is the power of, and philosophy behind, BC – ‘curve fitting’, to establish which options will potentially deliver maximum profit in a given situation. Markets tend to change rhythm (‘character’) so what works well today might not necessarily generate good signals three months from now. Anyway, if you want some ‘rough’ parameter settings to help get you started, see Essay # 1.1 above. TA is a huge subject and nobody can do it justice in one post. I have been studying it for 4 years, and I am still not making money. If you want to be an accountant or a doctor, you need to study; TA is no different. (If you want to ‘get rich quick’, buy a lottery ticket. :-) The other posts referred to in my previous note contain references to TA primers and info, to help you get started. If you are serious about making consistent profit, you will need to learn the science of ‘trading’, in addition to TA. By this, I mean understanding the management of risk, and sizing your positions accordingly. When I first began my study of trading, I assumed that the ‘Holy Grail’ would be a system that maximised return while minimising risk. Now I realise that to increase return one has to in some way increase risk, and that the ultimate system is one that balances return, drawdown, income consistency, and so forth in a way that rests comfortably with a trader's temperament, financial objectives, and lifestyle. You need to know all of the above, and also be able to apply the techniques in a relentlessly disciplined manner. I read somewhere that approximately one person in ten survives in this game, largely because many fall prey to their own emotions (greed, fear, pride, disappointment, impatience, panic, ...) and hunches. You have to be very confident that your entry/exit system is capable of delivering a positive expectancy (i.e. greater wins than losses) under all market conditions, in order to remain positive after (say) 10 consecutive losses have cut your account from (say) $50,000 to $25,000. Psychology plays a huge part. I could go into more detail here, but I would end up repeating a lot of the material in the other posts. I don’t want to dissuade you from becoming a trader. It is totally possible to succeed, and by that I mean make consistent profit, perhaps in excess of 100% per annum, year in and out. It’s got to the perfect job – no boss, staff, debtors, creditors, inventory; work whenever you like, for only a few hours, anywhere in the world the Internet is accessible. I’m just trying to make you aware of the size of the learning curve, if you are really serious about getting there. Finally, I would like to point out that I discovered BC about 4 months ago, and only use it as an adjunct to my existing system. The way that I use BC is discussed in the other posts. If you want to ask specific questions about the other posts, then I (and no doubt, others) will do my best to answer them. Meantime, start with the parameter settings recommended in this forum, try running them against a selection of stocks, and ‘paper trade’ them for 6-12 months. I guarantee that the experience will prove extremely valuable.
If you have read some of my other posts, you will be aware that I only discovered Best Charts 2-3 months ago, and only use it as a ‘second opinion’ to my existing trading system. For the latter, I use a different charting package, because I use complex indicators that I have developed myself, that are not available in Best Charts. There are many variables involved in trading, and I believe that there is no ‘holy grail’ system. It is up to each individual to come up with a system that suits his temperament, lifestyle, financial goals, trading time frames, philosophy toward reward and risk, and so forth. This is a huge topic, and I can’t begin to do it justice in one forum post. There are no simple answers. Even though you say that reading is making you confused, there is no substitute for knowledge, and it is only by sifting through a variety of different viewpoints, and then conducting your own trial-and-error process, that you will eventually discover what works best for you. Hence I suggest that you start by performing your own research, and formulating your own approach. Here is a starting point – Free Technical Analysis material on the net – http://www.incrediblecharts.com/technical+analysis.htm http://stockcharts.com/education/ http://www.investopedia.com/ http://www.tradingday.com/c/tatuto/ http://www.tradertalk.com/tutorial/ http://www.esignal.com/education/likepro/archive/default.asp http://www.equis.com/Education/TAAZ/ Candlestick patterns – http://www.litwick.com/glossary.html http://www.candlestickforum.com/ Success rates of different Technical Analysis patterns – http://www.marketscreen.com/help/chartpatterns/default.asp?Num=152 http://www.recognia.com/reference/patterndescr.htm Ten good books on Technical Analysis you might like to consider buying – http://www.esignal.com/education/likepro/archive/0303/032803.asp Also, check out the stock forums on the net. There are lots of these - http://www.incrediblecharts.com/forums/messages/board-topics.html http://www.traderclub.com/discus/messages/18/ http://www.trade2win.co.uk/boards/ Also, try this – LOTS of additional links – http://www.sms-india.com/investment/17/stock-trading-forums.htm See also Essay # 11.1 below for a vast number of links to TA and chart pattern related material. Now I will attempt to specifically address a couple of the points in your post. Which indicators work best? In my view, all indicators are ultimately derived from OHLCV price action, hence any confirmation is diluted, because they are not really independent of each other. As Best Charts illustrates, what buy/sell signals are generated depends as much on the parameters that are supplied, as the indicators themselves. The best advice I can give is to understand what each different indicator type (trend following, momentum, volume-based, etc) is measuring, how it is measuring it, and what effect this will have on the signals. Then run your own back-tests across the time frames you want to trade (day/swing) in Best Charts, and choose stocks, indicators, and parameters that consistently give the best results. There is a lot of repetitive manual work involved, but if you really are keen to make a living solely out of trading, then hard work is a small price to pay. Support and resistance? These are imaginary horizontal lines at which prices frequently reverse. Resistance lines are upper limits that the stock has difficulty penetrating when trending upward, and conversely, support lines are lower limits at which downward trends seem to reverse. Do some research about trend lines and chart patterns, perhaps starting at http://www.incrediblecharts.com/technical+analysis.htm I quote the following from another post of mine – My experience is that the market changes rhythm, or ‘character’, over a period of time, so that an algorithm that generated profitable signals in the past is unlikely to do so indefinitely. Furthermore, the markets are driven by emotion and reaction to unexpected news, so any probability analysis will be at best mathematically approximate. So conservative capital management is imperative.
Don’t get me wrong. Best Charts is fascinating, innovative, user-friendly, invaluable and neatly implemented. The concept of optimizing (or ‘adaptively calibrating’) indicators, based on profitability testing is a brilliant concept. I only discovered BC a few days ago, but I’m so impressed that I’m already using it as an adjunct to my existing MetaStock-based TA system. I can’t believe that it’s freeware! However, the novel idea of dynamic optimization needs to carry with it a few caveats. We need to understand the nature of the tool, and the processes, that we are dealing with. The most important thing to realise is that the gain (or loss) percentages quoted by the optimizations represent the best possible return that could have been made, and can occur only if one had known in advance how to calibrate the indicators accordingly. For example, let us assume that the optimized RSI for the last 88 days is a 5 day RSI, with buy and sell signals being generated whenever this crosses a 7 day MA of itself. But the same 88 days worth of data was not available to BC on the days that the buy and sell signals, which generated this maximised return, actually occurred. On each of those days, BC would have been looking at a different (earlier) 88 day window, resulting in completely different optimization, indeed one which might not even have generated a signal. In other words, the optimizations are all being performed with the benefit of hindsight, and hence the returns quoted are a maximised, rather than a realistic, value. But understand this – the above in no way undermines their usefulness. If we assume (and I would suggest that it is a reasonable assumption), that (1) each stock price action has its own distinctive ‘character’, or cyclical rhythm, (2) that this character is ever-changing (‘evolving’), and (3) that the most recent character has the greatest probability of imminent recurrence, then it follows that dynamic optimization is an smart and logical way of capturing, and then applying, this behaviour. So although the returns quoted should not be used as a realistic measure of past trading performance, the results of the profitability back testing do – in the context of the immediate future – represent a perfectly valid and objective means of (1) effectively calibrating these indicators, and (2) performing a head-to-head comparison of the potential value of each indicator, relative to the others, given the most current OHLCV data. But with one proviso (see below). First, an aside. There are other reasons why the returns quoted should not be taken as indicative – (1) The system makes no use of stop losses, or any premature full or partial exits used to progressively extract profits. In this regard, then, profits quoted by BC would likely be understated. (2) The system also makes no allowance for transactional (or ‘dealing’) costs, which would potentially have the opposite effect, i.e. of overstating profit on every trade. Given that BC is compounding gains and losses, both of these effects would be inflated exponentially. The more trades that occur over the period being back-tested, the more frequent the dealing costs, and the greater the effect of inaccurately inflated profits. There are situations where BC optimizes an indicator to two very close calibrations (whose crossovers generate the buy/sell signals), e.g. an EMA(1) and an EMA(2) for the BBI, resulting in a large number (e.g. 50 or more) of trades. Given that the transaction cost occurs once per trade, then, when the head-to-head comparison of indicators is being performed, allowances need to be made for the number of trades. Hence, I believe that the ‘average percent gain per trade’ column should be read as being a more valid basis of comparison than the ‘total percent gain’, for each indicator. I assume that the optimization process simply runs a set of values through each parameter associated with the indicator being optimized, and then calculates crossovers, buy/sell signals, gains (or losses) for each trade, and then compounds these for a total gain. Then this process is repeated for the next value, or combination of values, in the set. For example, to optimize RSI, BC would run the values 1,2,3, etc to a preset maximum through the RSI period, and then cross these with the values 1,2,3, etc through the MA being used for the crossovers. Thus the value pairs (1,1), (1,2) (1,3) … (2,1) (2,2) (2,3) … (3,1) (3,2) (3,3) … etc would be tested, and the compounded profitability calculated for each pair. The pair that generates the maximum profit becomes the optimized value. Of course I may be wrong, but that is my assumption as to how the optimization process would most likely operate. Now for the aforementioned proviso, which also concerns the disregard of transaction costs. (This leads to the subject of an earlier post of mine on this forum). If the ‘maximum profit’ for all trades across the back-test period is used to determine the optimum RSI value, then this is not a true optimization. My view is that the average precent gain per trade would give a more realistic value. Note that I am not repeating myself here. A couple of paragraphs back, I stated that the AVERAGE percent gain per trade was a more valid basis of head-to-head comparison of the indicators. Now I am contending that this is also a more valid basis for calibrating the indicators during the optimization process itself. Hopefully an example will clarify this further: It follows that, when evaluating the chart-based output generated by BC, I tend to disregard the indicators that whipsaw in and out of trades every 1-2 days. For the reasons I have stated, I see this data as being aberrant and of little value. Finally, some general caveats that don’t specifically relate to BC. The fact that all indicators are ultimately derived from OHLCV values undermines, to some extent, any confirmation effect of coinciding indicator signals, especially where the indicators concerned are calibrated across a similar time frame. Offsetting this is the fact that BC uses a diverse smattering of price, volume, and momentum based indicators, all summarised into BBI, which is then itself optimized. So common sense should be used, to decide exactly how strong the confirming effect is. Indicators are capable of forecasting the probable direction of a trend, but not necessarily its gradient, or future duration, which ultimately determine profit (or loss). Like other TA systems, BC will generate effective entry signals, but even the best entry signals offer no guarantee of profit. In general, an earlier entry will net one a better price, but with lower probability that a confirmed trend has been established. Profit is only achieved if the trend persists long enough to allow costs to be overcome, and exit occurs before the profit is consumed by the eventual reversal. Moreover, of the factors that ultimately determine one’s bottom line profit (e.g. entry point, exit point, stock selection, position size, etc), position sizing way outweighs the others in its significance. With casino-type games of chance, no back testing is necessary, because the probabilities generally remain constant, and can invariably be precisely calculated (e.g. drawing into an inside straight in Poker, getting a man in off the bar in Backgammon with 3 inner points blocked). But the money markets are driven by crowd emotion (fear and greed), and also shorter term price convulsions resulting from impulsive or erratic trader behaviour caused by unforeseeable political or economic events. Hence one is dealing with probabilities that are at best, mathematically approximate; no amount of back-testing, however exhaustive, can ever be totally predictive. This essay is in no way designed to undermine what is one of the most innovative and valuable charting products I have ever discovered. I love Best Charts! I have made some assumptions as to how BC operates, and, if these are incorrect, then I leave it to the reader to mentally make any necessary adjustments to the points I have made. I’m hoping that my commentary helps users of BC to better understand the benefits and caveats inherent in dynamic optimization (aka adaptive indicator calibration), and therefore how to best interpret, and exploit, the statistics output by BC. Any allusions to shortcomings are in no way intended as a criticism of the product itself. Re AIO and BT, see also the discussion in the following threads: http://www.stock-anal.com/ubb/Forum1/HTML/000901.html http://www.stock-anal.com/ubb/Forum1/HTML/000910.html
Indicators are frequently late because they are derived from OHLCV price action – the OHLCV has to exist before the indys can be calculated. Price based indicators like MAs must lag because they summarise prior OHLC values. Oscillators like RSI can appear to lead because they measure rates of change that can point to future direction. However, I find that, as a general rule, ‘leading’ or early signals are less reliable, because they point to trends that are not yet established. You will enter at a better price, increasing amounts won and decreasing amounts lost for each trade, but the lack of reliability means that your win/loss ratio will be poorer. It is a finely judged compromise. Indicators can be useful where they highlight or summarize data that might not be immediately obvious by simply viewing the raw OHLCV data. The degree to which many indicators lag depends on their calibration, e.g. a 25 day MA lags more than a 9 day MA. I would have thought that BC’s optimization process, in finding the combo of parameters that maximizes profit, would have ‘tightened’ the parameters in a non-trending market, allowing earlier entries and exits. In other words, it seemed logical to assume that, given its self-adjustment capability, BC would shine above other ‘conventional’ products in difficult markets. However, judging by the earlier posts in this thread (including Rik’s) one might conclude that this is not so. However, much would depend on the time frame used for the back-test, and the allowable range of parameters for each indicator that get crunched by BC during the back-testing. As I have said in earlier posts, the idea that the most recent price action has the greatest probability of imminent recurrence, makes sense to me. However, until I see the results of an independent profitability back-test of optimized versus unoptimized signals, across a huge sample, I will keep an open mind as to the extent, if any, of the benefits of optimization. Your comment on BCI is interesting, and suggests that BCI might have been developed specifically with the optimization process in mind. However, assuming that it must in some way be derived from OHLCV, then I see no reason why it should be more receptive to optimization than the other indicators. I haven’t tested it comprehensively, but it does appear that it frequently gives earlier signals, which I suspect would work better, on average, in a sideways market. If BCI can be optimized to give maximum profitability by generating buy/sell signals on certain dates, then it stands to reason that the other indicators, if back-tested across a wide enough range of parameter combos, should be able to generate signals on these same days, and hence be able to compete with BCI in the context of quality optimized signals. Having said all this, BC allows any combo of both fixed and optimized indicators to be created in opd.txt for any stock (see document referred to in post 1618 for details), so one can always use one’s own optimization algorithm, if absolutely necessary. One final point. The simplest way to devise a system of entries is to draw (visual) trendlines between local highest and lowest points that coincide closely with your chosen trading timeframe (intraday, swing, longer term, whatever). Then you have to ascertain the costs involved – dealing costs (brokerage, spread, slippage, whatever), and also the trend establishment and reversal cost, i.e. how many price bars, on average, follow the peaks and troughs before you are satisfied that a trend has been SAFELY established, and then (for the exit part of the deal) eventually reversed. Then you can manually calibrate your system to give entries at these points. Now if, on average, the distance between the highs and lows (i.e. revenue) exceeds all of the costs, you will ultimately profit; if not, your system is incapable of overcoming the prevailing market conditions. All this of course assumes the past is a reasonable approximation of the future, a premise that lies at the core of all technical analysis. My point is this – it doesn’t matter what indicators (or channels, candle patterns, whatever) that you use, the important thing is that they are calibrated to coincide, as consistently as possible, with the optimum buy/sell points in your preferred trading timeframe. ===== END OF POST ===== [This message has been edited by david_louisson (edited 12-12-2006).] IP: Logged |
|
tommymacd Member Posts: 1 |
posted 08-20-2004 01:23 AM
I have been doing virtually the same thing that Alvin has recommended. I have used this software intermittently for the last 6 months. Paper trading indicates a return of almost 25% based on outside research. In the last 3 days my paper trades have produced a 29% ROI. Not too shabby. I'm convinced. This is a great tool.... IP: Logged |
|
david_louisson Member Posts: 303 |
posted 08-20-2004 09:27 PM
ESSAY # 2.1 ===== CHANGES IN VERSION 4.33.2 ===== Where relevant, the following supersedes information in the ‘DISCUSSION ON PARAMETER SETTINGS’ section. There is now a menu option ‘Stocks/Options...’ which allows entry of the following – Option Sample value Stock Markets: UK Stocks These replace previous menu options on the ‘Stocks’ drop down menu. Settings are saved between sessions in the file ...\M-C\options.txt This means that it is no longer necessary to re-enter these settings each time one runs BC.
Versions 4.5 and 4.6 add the following new chart types: Heikin-Ashi candlesticks These can be accessed by clicking the 'MORE' button on Best Charts's toolbar. For more information on how to use and interpret these new charts, please see: For more information on the T3 indicator, please see my comments in the following thread:
[This message has been edited by david_louisson (edited 11-17-2006).] IP: Logged |
|
david_louisson Member Posts: 303 |
posted 10-01-2004 04:57 PM
ESSAY # 3.1 ===== USING HISTORICAL QUOTE FILES ===== Essentially you simply enter the symbol name, and click the ‘H’ button, which will cause BC to display historical quotes from files you have previously downloaded, or otherwise created, on your PC. This approach can be used to allow back-testing of selected periods of data, i.e. by editing the file to leave only the desired date range intact. There are two issues – file location, and file format. I will deal with each of these here. FILE LOCATION The default folder is C:\M-C\HIST, so the simplest way is to make sure that all of your historical quote files reside there. Each file should have a .CSV extension. Then you type the name of the file, without the extension, into the symbol field, and click the 'H' button. You can change these default settings, as follows. 1. Use 'Stocks>... Set Historical quote file path' as follows: type C:\[full folder path]\ in the first box (don’t forget the final backslash) 2. You must tell BC to 'Use specified file path'. Otherwise it will look in the default C:\M-C\HIST. 3. Type in the symbol name and click the 'H' button. Example – A word about non-US exchanges – The key is this: the folder path + the symbol typed + X + the suffix/extension Where X = the Yahoo suffix pertaining to any exchange that you select from the dropdown. (For USA, this is null.) The absence of a file causes the following error message to occur:
This should be as in the following example – Note the following carefully, as an invalid format can cause BC to close/crash. (1) The presence of the header record (apparently it is optional, but I believe that it is safer to have it, and it enhances readability).
[This message has been edited by david_louisson (edited 11-28-2004).] IP: Logged |
|
hexachord unregistered |
posted 10-06-2004 09:56 AM
Dear David; I am the one who found 88 data samples to be the most useful. That said, I have found, for whatever reason, it could be because of improvements made to BC or improved data from "Sources 1, 2, or 3" that the program is now able to manage (successfully AIO)larger amounts of data. If I could resubmit my original article (which says more or less what you nicely said in your article) I would suggest that BC users select under After running AIO on 252 days of data, I use the indicator that generates the highest average return per trade AND trades at least 16 times (sometimes I will go with a signal that traded only 14 times in the last year if the returns are significantly better and/or more consistent. Qualifying stocks, for me, are those that after AIO, with BT set to 252, produce an indicator that had an average return per trade of equal to or greater than 8% and 16 or more trades. Like you, I use this program for swing trading exclussively, however, it seems that if one wanted to invest for longer periods, one could use weekly data rather than daily data instead. I've thought about it, but I just don't trust companies enough to hold their stock for longer time frames. Dr. Luke Palmer IP: Logged |
|
hexachord unregistered |
posted 10-06-2004 10:05 AM
Dear David; I like your suggestion for the Portfolio tool. I neglected to mention that the success I have with BC is in large part due to the fact that I only rely on BC's BUY signals and rarely wait for SELL signals, I typically sell with only a few percentage points in gains and then move my money into another stock that is generating a BUY signal. A percentage point or two a week happens to generate plus or minus 100% gains over the course of a year. Dr. Luke Palmer IP: Logged |
|
david_louisson Member Posts: 303 |
posted 10-09-2004 03:37 PM
Dear Dr Palmer Thank you for sharing aspects of your approach regarding optimum BC use, and trading. It is interesting that your comment regarding the use of 88 versus 252 data samples (‘quotes’) coincides with BC’s release of their Multi-AIO facility. As I understand it, Multi-AIO allows the user to enter up to 8 different values for ‘number of quotes’, and have BC test which one delivers greatest profit. Thus the optimum ‘Number of quotes’ is determined, an AIO is used performed using this number, and then a TA of the resulting charts is displayed. I’m not exactly certain how BC decides which indicator to use in determining the OVERALL optimum, e.g. for a certain stock RSI might deliver greatest profit when back-tested over 60 quotes, while MFI might deliver its greatest profit when back-tested over 150 quotes. Presumably it would use BBI, the weighted summary of all indicators? Perhaps Mr Admin would care to comment. I totally concur with your point that it is possible to make fantastic annual returns by compounding small weekly percentage points. For example, 1% per week compounded across 52 weeks = around 68% p.a.; 2% compounded per week = around 180% p.a., and so forth. Throw leveraged instruments (options, futures, CFDs, etc) into the mix, which also allow one to short the market, and we have the potential for enormous exponential growth. Which brings me to my most important point: position sizing. For me, the gain made on each trade is only a small part of the equation; it is the percentage of one’s total capital pool that one stakes, per trade, that has the single greatest bearing on bottom line. Establishing the maximum that can be SAFELY risked, i.e. finding a balance between return and drawdown that one is comfortable with, is the most challenging part of the exercise. Best wishes IP: Logged |
|
david_louisson Member Posts: 303 |
posted 11-09-2004 01:03 AM
ESSAY # 4.1 ===== INSTALLING NEW VERSIONS OF BEST CHARTS ===== Rather than uninstall old versions of BC, I rename the folder where the prior version resides to Then I re-install BC from the link http://www.stock-anal.com/best-charts.exe I don't know whether uninstalling BC removes the C:\M-C folder, but the above approach is guaranteed to retain it. This also has the benefit that any Desktop icon linked to BC will automatically point to the new version. As far as I can tell, there are two versions of BC, the trial one and the 'real' one. The former has the 30-day expiration message, while the latter simply looks for your password in C:\M-C\mkt.stk To summarise, there are two important conditions – If either of these conditions is not met, BC will not work. I have not tried to uninstall BC, just in case the C:\M-C\ folder gets deleted in the process. Apart from loss of the password, all stock and portfolio settings are also stored there. Moreover, it is perhaps a sound idea to make a back-up copy of the C:\M-C\ folder, especially if you have created extensive portfolios. I also maintain at least two versions of BC on my PC, e.g. [This message has been edited by david_louisson (edited 11-28-2004).] IP: Logged |
|
david_louisson Member Posts: 303 |
posted 11-09-2004 01:15 AM
The following is a transcript of a post of mine on another forum. Those who are struggling with their trading may find some of the points helpful. ESSAY # 5.1 Reading through this thread, it looks like you have applied most of the techniques used by the contributors. However, I would like to try to help – here are my ideas, for whatever they might be worth. Unlike many of the ‘conventional’ exponents of trading and TA, my system (which is still under construction) is based around what I call the ‘casino approach’. Casinos are guaranteed to profit long term because they observe four simple principles – 1. The only offer games that have an ‘edge’ in favor of the house. Let us now attempt to apply these principles to trading the markets. 1. EDGE (Trading equivalent = TREND FORECASTING, ENTRIES & EXITS) You must develop a system of entries and exits that delivers positive expectancy, i.e. total net winnings exceeds total net losses. The only way I know of achieving this is by running a back test across historical prices. Any kind of TA-based system makes the assumption that the trends of the past are more likely, than not, to recur in the future. Start by taking the price chart of a stock or index, and marking the highest and lowest points relative to the time frame (see point 2) that you want to trade. Then attempt to curve fit channels and/or indicators that give the best possible return through these points. This calibration is a trial-and-error process (using a charting package that allows for system testing). You also need to likewise calibrate the positions of your stop losses, so that the system as a whole delivers maximum possible profit. And don’t forget to factor in transaction costs – brokerage, spread, slippage, borrowing costs, whatever. Then try the same calibrations against different stocks and indices (given that properties like volatility and cycle wavelength are likely to vary from stock to stock), to get a feel for how consistently these operate, and modify accordingly. It is also a good idea to split your system testing across different ‘chunks’ (time periods) of the data, to convince yourself that the same algorithm works satisfactorily across different market ‘climates’ (trending, ranging, etc). Alternatively, you will likely find that optimum results are attained by using approaches tailored specifically for different climates. There are a daunting number of indicators, trendlines, chart and candle patterns, and other TA tools that can be used, but you could start with an already recommended approach that makes sense to you, and then vary, and optimize, it to suit your own trading time frame, and style. Try to use at least one indicator from each of these categories – trend following, momentum oscillator, volume based (keep in mind that, because all indicators are ultimately derived from the same OHLCV data, using too many indicators of the same category can undermine apparent confirmation). Anyway, the more effort put into the trial-and-error process, the better the chances of achieving an optimum result. You can also bring other semi-independent variables into the mix – news and fundamentals; market breadth: how other stocks in the same industry sector are performing, both nationally and overseas; also national and international indices. All of these can give further backing as to the most likely direction of your candidate trade. The old maxims of trading with the longer term trends, and exiting losers quickly while letting profits run, are also sound. Where possible, try to bring these variables into your back testing, and gauge the extent to which they filter entry/exit signals that affect the result. Analyze the losing trades, and try varying the parameters to eliminate them. However, examine the effect that this elimination has on the winning trades. Forecasting can be a game of fine edges – sometimes patterns that lead to winners in some scenarios lead to losers in others. Try to find out if there is supporting evidence why. A further reason to back-test is to allow you to hone and understand these probabilities, and gain a feel for the prevailing rhythms in the underlying price action. Bear in mind that casino games offer precise odds, while the market probabilities – driven by fear, greed, and crowd response to news – can, at best, be calculated approximately. Hence to trade with a degree of confidence, you ideally require an expectancy that is SIGNIFICANTLY positive. Without this level of confidence, you will find it more difficult to adhere to your system should you encounter a series of losses. It is easier to forecast trend direction, than its future gradient, or duration (Elliotticians might disagree!) Obvious as it is, always remember that the bottom line is that if, on average, the prices ‘trend’ sufficiently to overcome costs, and that the positions are closed before the eventual reversal erodes the profit, you must ultimately win in the long term. Otherwise your system is ineffective across the climate being tested, forcing you to sit out of the market. It is both as simple, and as difficult, as that. 2. HIGH TRANSACTION THROUGHPUT (Trading equivalent = FREQUENT, SHORT TERM TRADING) To achieve this, you ideally need to be day or swing trading, with an instrument (e.g. CFDs, options) that allows both long and short positions, diversified across several independent, simultaneous trades. Note carefully that (unlike the casino) the independence factor is undermined by the fact that all markets tend to move in unison, to some extent. CFDs not only allow huge leverage, but also hedging, that can be used to alleviate risk. One possible approach is to take a number of concurrent long positions on stocks or indices that have just begun to trend upward, while simultaneously taking short positions on a number of stocks that have just begun to trend downward. That way, the overall market direction becomes less significant. High transaction throughput can also help with the psychological aspect, as (given a mix of long and short positions) it is unlikely that all trades will fail simultaneously (and hence with CFDs, cause a margin call). In this context, it is important to size positions inversely proportionally to volatility (i.e. smaller position sizes for more volatile stocks). You might also like to experiment with adjusting position size to reflect the number of ‘external’, independent factors (see previous section) that are backing the candidate trade. Concurrent positions require extra management time and effort, but my view is that the serious trader should be willing to do whatever it takes to deliver the optimum result. Used judiciously, diversification can be used to leverage time, increasing return without severely altering risk. As in the old adage, there is ‘safety in numbers’. Finally, higher transaction throughput provides a faster growing, more statistically significant sample size for testing purposes, and accelerates the learning process. 3. CAPITAL RESERVES (Trading equivalent = POSITION SIZING, MONEY MANAGEMENT) This requires a thorough understanding of return and risk. In my view, there are four important bottom-line elements in measuring system performance – Mathematically, we need go no further than this. In theory, the best system is the one that maximizes annual return without risking irretrievable drawdown. There are plenty of Monte Carlo simulators available to test return-versus- drawdown for entered win and loss back-test stats, allowing you to estimate optimum position sizes. However, life dictates that there are also practical considerations. These are – When I first began my study of trading, I assumed that the ‘Holy Grail’ would be a system that maximized return while minimizing risk. Now I realize that to increase return one has to in some way increase risk, and that the ultimate system is one that balances return, drawdown, income consistency, and so forth in a way that rests comfortably with a trader's temperament, financial objectives, and lifestyle. The basic rule-of-thumb is: Increased position size = Increased return = Increased drawdown. If in doubt, size your positions modestly, and be content with the lower return, sleeping more easily in the knowledge that your trading account will live to fight another day. 4. RIGID GAME RULES (Trading equivalent = disciplined adherence to your mathematically-proven system) This speaks for itself. However, it’s possible to have mastered all of the theory, but falter as soon as the stakes become real. I have no answer to this complex psychological problem, other than to have a trusted friend place your trades for you, if this is the case. GENERAL COMMENT My view is that it should take 5-10 years to become a consistently proficient trader. The first 1-2 years should be spent in constructing your system, and then paper trading to prove the theory in practice, to attain a degree of confidence. Belief in your system is everything. Lose faith, and despondency will eventually affect your trading decisions, causing deviation from the statistically proven approach, and ultimately making the hours of back-testing meaningless. Given that the ultimate potential is a lifetime of financial freedom for you and your children, the preparatory spadework is worth the effort. The longer your trading time frame, the more historical data you need to execute a statistically significant back test, and also the longer the time needed to paper trade through the longer cycle wavelengths. I think you probably need to start by re-visiting point 1, putting your failed trades through a visual back-test. Are they of a statistically significant number that some patterns can be detected? For example, are you entering or exiting positions too early (= higher throughput, and higher return if successful, but lower probability of success), or too late (lower throughput, average return per trade; higher probability)? Missing significant trends (requiring excessive confirmation for ‘setup’ criteria, or signals that are way too late)? Are you setting your stops too close (getting stopped out of potential winners prematurely, too often), or too distant (giving back too much of your hard-earned profit on losing trades)? Is your strategy working for some stocks, but not others? In a trending, or ranging market? Or have you just been unlucky (need a higher transaction frequency, to allow the positive expectancy to shine through)? If you assemble enough data, you should be able to discern patterns from the outcomes. That is a good starting point. This is a huge topic, and what I have attempted to cover is but the tip of the iceberg. SUMMARY OF THE FOUR RULES 1. The 'System' – Develop, back-test (and refine) a system of entries and exits that delivers positive expectancy (i.e. overall wins exceeds overall losses) across all phases of the market. 2. High throughput – Diversify by taking uncorrelated, simultaneous positions to increase return disproportionately to risk. 3. Capital and risk management – Use stop losses, and manage capital conservatively. Preserve your capital to ensure that you will be able to continue trading tomorrow. 4. Psychological – Apply all of the above mechanically, with DISCIPLINE and PATIENCE.
1. DON’T BE GREEDY 2. REMAIN POSITIVE 3. BE CONSERVATIVE 4. REMAIN UTTERLY DISCIPLINED 5. GENERAL REMINDERS
* Trade in the direction of the longer term trends. * Trade with independent confirmation from other markets, sectors, stocks. * Apply the assumption that the current trend will continue: * Trade what you SEE on the chart, NOT what you think or feel
* GREED (can cause overtrading) * FEAR (anxiety can cause people to exit winning trades too early, reducing profit) * PRIDE (unwillingness to admit failure can cause people to hold onto losing trades too long, increasing losses) * IMPATIENCE (can cause 'position-hunting' through hunches and guesswork) * DESPONDENCY (can cause irrational attempts to recover losses by excessively increasing position size) * COMPLACENCY (can also cause imprudent increases in position size)
Every successful trend-following system somehow seems to incorporate elements of the following method: 1. Setup and stock selection: Buy into a strong upward trend, immediately it resumes itself following a short-term pullback. Maxim: trade in the direction of the prevailing trend (to maximise both the probability of a win, and also the potential size of the subsequent price movement). 2. Timing of entry: earlier = higher risk of being stopped out, but greater profit (more favorable entry price) if successful. Excessively late risks missing the trend altogether. 3. Distance of stoploss exit: tighter stop = higher number of trades stopped out (lowers win rate), but keeps average loss smaller. Maxim to aim for: cut losses short. 4. Distance of profit exit: looser trailing stop = lower risk of profit being cut earlier in the trend, but gives back more profit when ultimately triggered. Maxim to aim for: let profits run. 5. Position size: the ultimate determinant of bottom line. Priority: manage risk. Apply capital so that risk is approximately equal (maximum 2%) across each trade undertaken. Keep sizes small enough to avoid emotional involvement. 6. Psychology: trade a proven, pre-defined plan consistently, with patience and discipline. Trade what you see on the chart, not what you think or feel. Patience means being willing to suspend trading when market conditions are unsuitable for one's system (e.g. sideways market). There is sufficient randomness in short term movements to make minor variations in execution largely irrelevant, e.g. which indicators are used, variations in timing. However, it is important to prove your own variations to your satisfaction, so that you can confidently trade your plan.
a) Buy the strongest performing stocks in industry sectors that are outperforming the market index (use "relative strength comparative" or similar indicator). b) Trade in the direction of the market index's trend (to reduce "market risk"). c) Buy stocks whose TA exhibits the weakest and/or most distant areas of potential resistance (i.e. best reward to risk ratio, greatest freedom of price movement). d) Buy stocks that are liquid enough to avoid slippage. e) If you follow fundamentals, buy stocks exhibiting the best fundamentals. f) As an alternative (or foil) to #1, also buy stocks breaking out of a sideways pattern (or narrowing wedge), with high volume. g) For short positions, the reverse applies, but in general it's necessary to act more swiftly, because bear movements tend to be sharper.
h) Consider diversifying capital across multiple uncorrelated stocks that meet the above criteria, to reduce loss, should one trade "earthquake" against you. Benefits of diversification and/or high throughput include: mitigation of risk, greater opportunity for "edge" to prevail, smoother consistency of income, accelerated learning curve, more frequent compounding of gains (and losses!), lower emotional involvement, accelerated statistical confidence. You will save yourself time and MONEY if you heed the following: You can NOT predict which way the market will move in the future. The best that you can do is surf the strongest trends, based on the probability that they will continue.
When analyzing the profitability of your system, there are two essential factors: a) the number of winning trades to losing trades (e.g. a value of 60% means 6 wins out of every 10 trades, on average); b) the average win size to average loss size (e.g. a value of 1.7 means that the average win is 1.7 times the size of the average loss) Note that wins and losses are calculated after deducting costs. Note also that combining (a) and (b) gives the profit factor (PF), "expectancy" or "edge" (call it what you will) A good (and realistically attainable) target is to aim for > 50% value of (a), and > 2 for (b). The latter means you are choosing trades where the profit potential is > 2 x the distance from entry to your stoploss. I've brought this up because I want to discuss some of the biggest issues surrounding entry and exit, irrespective of trading time-frame, market traded, instrument used, indicators used, etc. Irrespective because the bottom line is that you must necessarily enter on a price bar and exit on a price bar. Provided that your method involves some kind of trend following, all of the following points apply:
The earlier you enter, the higher the risk, in the sense that a "genuine" reversal might not have occurred, i.e. it is just a minor deceleration ("noise") along the trend. As compensation, the earlier entry will generally attain a more favorable entry price. Hence an earlier entry will, on average, increase (b) but decrease (a). In other words, your profit will be greater on the occasions that you do win, but you will win less often. This gets back to your original question about divergences. Every trend must necessarily decelerate before reversing, and divergences will tend to pick up these decelerations. But not every deceleration will actually result in a profitable reversal. Hence the higher risk. Conversely, a later entry will, on average, increase (a) but decrease (b). The later entry increases the probability that a profitable trend is emerging; however, one is forfeiting eventual profit at the start of the trend, by entering at a less favorable price. As I've said elsewhere, profit is determined not by WHETHER the price moves in the anticipated direction, but HOW FAR (and how quickly). Profit is attained only if the price moves far enough to overcome costs (see point 3 below). Because it is difficult to predict how far price will move (some traders attempt to use previous support and resistance, others Fibonacci retracements or space/time ratios, as a guide), if the trend turns out to be short lived, then delaying entry too long actually results in increased risk.
It is widely accepted that the exit is more important than the entry. Here are some major reasons why. If you set your stoploss too tight, you will get prematurely stopped out of trades that would ultimately result in a profit, more frequently. This is a double whammy: suffering a loss, AND forfeiting a potential profit. However, the tighter stop means that your losses will always be small. This results in an extremely poor (a) value, but a high (b) value. The same applies to trailing (profit-taking) stops. A tight trailing stop will guarantee locking in some profit, but has the potential to prematurely curtail extended future profit. In other words, it can "cut winners short". Of course, the converse applies if stoplosses, and trailing stops, are set more loosely. The "looser" the stop, the less frequently premature stop-outs will occur, but the greater the average loss (or forfeiture of profit) becomes, when the stop is triggered. Hence the value of (a) increases, but at the expense of (b). Another major reason why stop-setting is crucial, is its effect on fixed fractional position sizing (the method that most traders seem to use). As an example, let's say your trading account balance is $10,000, and you are willing to risk 2% (the recommended maximum) on a given trade. 2% of $10,000 is $200. Let's suppose that stock XYZ gives a "buy signal" at $2.00, and you decide to set your stoploss at $1.90. Then if the stop gets triggered, you will lose 10c per share purchased. Given that you are willing to put $200 at risk, then you would buy 2,000 shares (i.e. 2,000 shares @ 10c loss per share = $200 loss). However, if you were to set your stoploss at $1.95, and it is triggered, then you will lose 5c per share purchased. Given that you are willing to put $200 at risk, then you would buy 4,000 shares (i.e. 4,000 shares @ 5c loss per share = $200 loss). In other words, the tighter the stoploss ($1.95 vs $1.90), the more shares purchased (4,000 vs 2,000), and the more capital "tied up" in the transaction (admittedly an irrelevance if one is trading derivatives on low margins). Assuming that one works to the above formula, the loss is the same ($200 vs $200), but the potential for return is doubled (4,000 shares x favorable price movement vs 2,000 shares x favorable price movement). Offsetting this is the greater likelihood that the stoploss will be triggered (as already discussed). An alternative to using a trailing stop is to extract profit by closing positions "progressively", by selling off parts of the position immediately arbitrary target points along the way are reached. The advantage to this approach is that the profits taken are totally secure, whereas stops are more vulnerable to gaps and slippage, especially should a position suddenly "earthquake". The drawback is that potential profit is being sacrificed prematurely, should the trend proceed to "mature" a great deal further. One final point: some traders don't use "mechanical" stoplosses. But to do so requires both the judgment to recognize when a trade has "gone awry", and the discipline to exit immediately at that point.
(i) Transaction costs: brokerage, spread, slippage, financing costs, etc. (ii) Profit forfeited between the trough/peak and the entry point (determined by how early/late the entry is – see point 1 above). (iii) Profit consumed between the trough/peak at the end of the trend, and the exit (determined by how tight the stops are – see point 2) If you can (on average) overcome all of these costs, you will profit; otherwise you will lose. That is the bottom line.
All of the above applies to any trend following system. Whatever your method, you should BACK-TEST it across a sample (large enough to be statistically significant) of historical data. Here's why: (i) The overall results will give you values for (a) and (b), telling you how THEORETICALLY profitable (or otherwise) your system is. This is especially important if you plan to be a "mechanical" trader (i.e. entries/exits follow a set of iron-clad mathematical rules). (ii) A study of the individual trades will help you gauge whether you are (on average) entering too early or too late, and whether your stops are too tight, or too loose. If you plan to be a "discretionary" or "intuitive" trader (i.e. entries/exits follow a set of general principles, applied with judgement), developing this kind of "feel" for what is likely, or unlikely, to happen in different situations, is absolutely vital. (iii) Study the losing trades, and find out the reasons why. Vary your parameters, and see their effect on (a) and (b). What effect does eliminating/alleviating the losing trades have on the winning ones? Find the best compromise. (iv) The most important reason to back-test (as if the above were not enough) is CONFIDENCE. You must believe that your system will continue to work, even during the rough times, if you are going to continue applying your system with DISCIPLINE. This is even more important for a discretionary trader, whose more "flexible" rules provide greater temptation to deviate. (v) It's important to test different samples across differing market conditions (e.g. bull market, bear market, trending, ranging, volatile, inert). This will help tell you over which conditions your system performs best, and what adjustments need to be made to cope with the differing underlying "climates". One caveat to back-testing. Unlike casino-based games of chance, the "markets" (in reality, crowd behavior driven by – amongst other things – greed, fear and other emotions) do not run according to precise probabilities. What has happened in the past is only APPROXIMATELY more likely, than not, and in some APPROXIMATE form, repeat itself in the future. So back-testing results are guidelines, no more. But if there is any non-random behavior amongst the chaos, then it can be mathematically approximated, and (God willing!) profitably exploited.
Putting the math aside for some real-life practicality: Some traders are willing to live with a system that delivers a value of (a) < 50%. As we have seen, we can time entries and set stoplosses to balance (a) and (b) to suit our personalities, goals and lifestyle. To deliver long term profit, the product of (a) and (b) must deliver positive expectancy. In other words, if one is willing to let (a) slip as low as 40%, then (b) must exceed 1.5, and so on. However, if trading is one's primary source of income, then there are other important considerations. The first is cashflow. A low value of (a) could mean a long wait for the next big winning trade, necessitating careful household budgeting. The second is psychology. Again, a low value of (a) could mean a long wait for the next big winning trade, leading to any of impatience, despondency, lack of confidence, and ultimately temptation to deviate from the system's rules and principles.
If you have become fearful about entering trades, consider the following: 1. REALISTIC EXPECTANCY 2. STATISTICAL CONFIDENCE 3. MANAGE RISK 4. POSITION SIZE 5. MARKET "DISINFORMATION" 6. MENTORING
From what I have discovered from my own trial-and-error processes, short term "swing" trading (e.g. positions closed within 3-15 days) requires a different approach than "position" trading (e.g. where trends can last 30 days upward). Unlike you, however, I haven't seriously looked at intraday, because my data source provides only EOD bars. Short term trading is more difficult, but (if you can master it) offers significantly greater profits. Some possible reasons follow.
1. Transaction costs are larger, relative to price movement. The swing trader might ride a price movement from $2 to $2.05, while the longer termer from $2 to $2.50, but both are paying (virtually) the same transaction cost. 2. Transaction costs occur more frequently: they hit you every time you enter, and exit, a trade. Short term = more entries and exits. 3. Prices tend to "trend" more in the longer term. Short term movements are more random. The shorter the term, the more critical entry becomes. 4. Extending the idea in 3, sudden gaps and spikes are mere ripples ("noise") to the longer termer, whereas they can have a massive impact on the swing trader, even shaking him out of a trade (since his stoploss must necessarily be tighter). 5. Unless the swing trader is analyzing intraday bars, his analysis will have less data to work with. Most indicators have been designed to operate with larger calibrations (e.g. RSI's "factory default" is 14), tailoring them for longer term trading. My research suggests that simply lowering the calibration (e.g. using an RSI(5)) for short term trading isn't really that helpful, because the indicator has less data to work with, and one day's aberrant movement can severely distort the result.
6. You get to compound gains (and losses too, of course, but we assume a system with overall positive expectancy) more frequently. (1% per week compounded = 68% per year; 1.5% per week compounded = 117% per year). 7. If your system of entries and exits is delivering this positive edge, then high throughput gives the best and fastest opportunity for the edge to apply itself. Points 8 onward are logical outcomes of this phenomenon at work. 8. Your learning curve is accelerated, because you acquire a greater "mental database" of trades more quickly, along with your experience at managing them. 9. For the same reason, you attain statistical confidence more quickly. As an example, 70 wins from 100 trades engenders vastly greater confidence than 7 wins from 10 trades. 10. Smoother income, and equity curve. As an example, "Mr A" makes an average of one trade per month, "Mr B" makes 3 trades per week (assumes some diversification). A "losing streak" of 5 trades means that "Mr A" has no income for 5 months. 11. Psychological benefits, in that higher throughput means that one is less likely to become emotionally involved (i.e. dependent) on the outcome of one trade. That helps to engender a "correct" view, that trading is a game of (albeit fuzzy) probabilities, and that frequent losses are very much a part of it. Hence you become equipped to overcome fear of loss more quickly, permitting a greater level of "detachment". 12. Consider using uncorrelated diversification. Mitigates risk, and helps to retain some profit should only one of the multiple positions "earthquake" simultaneously. 13. The short term trader can profit in what the longer termer would call a "sideways" market, provided that the sideways channel is wide enough to cover costs and deliver a reasonable profit. In other words, increased potential trading opportunities.
For better or worse, longer term trends (e.g. a 250-day MA) are less significant to a short term trader, as he seeks to catch both short upward and downward movements WITHIN such longer term trends. However, it is nonetheless vital to trade with the trend (for a swing trader, a 25 day movement summary might be more appropriate), because profit is determined by the extent of price movement, not just the forecast direction. The swing trader is reliant on squeezing every last drop of profit out of each trade, because he is dealing with small numbers, but more frequently. Fundamental analysis is also less relevant because these types of corrections in share value are more likely to be played out over several weeks, rather than a few days. Hence the swing trader must rely on technicals alone. Indicators are also less helpful. See the reasons in point 5 above. The swing trader needs a strategy to overcome the possibility of overnight gaps, and also the higher level of volatility RELATIVE to potential "profit exit", due to the short timeframe. Stoplosses must be set distant enough to overcome this volatility, which undermines the "cut losses short" principle, and consequently the potential "size of average winner" to "size of average loser" ratio (Van Tharp's "R" value). Hence it is important to attain a higher "#wins" to "#losses" ratio, to compensate. Perhaps one should consider locking in profit by progressively closing down part of the position, rather than using trailing stops, which can be gapped over. Note that this undermines the theory of "letting profits run", but we are trying to nail down risk here, even if potential return is compromised. The apparent violation of two trading maxims highlights the different nature of, and difficulties involved with, short term trading. The principles of "letting profits run" and "cutting losses short" are vital to longer term traders, because these are trend following assumptions, and prices tend to trend more in the longer term. (You cut losses short because, when the market moves against you, the assumption is that the unfavorable trend will continue. You let profits run because the "law of trending" means that, on balance, the favorable trend will continue.) I am NOT saying that one should behave oppositely to these maxims, e.g. by averaging down, but that (i) in the lower trending nature of the short term, these principles will lose much of their effectiveness, and (ii) one must understand the reasons behind rules and principles, and apply them correctly in the relevant situations. I've not gone into too much detail, because (i) exact entries and stop placements would depend on the characteristics of the stock, instrument, etc being traded, and (ii) I'm still in the process of trialling my own short term system. Logic would suggest that the bottom line is that any system of entries and exits will work, provided that it is geared to exploit statistically significant non-random price behavior, and is capable of overcoming costs in the process, on an "on balance" basis ("on balance" highlights the need for conservative position sizing).
[This message has been edited by david_louisson (edited 01-03-2006).] IP: Logged |
|
david_louisson Member Posts: 303 |
posted 11-09-2004 01:45 AM
To the users of Best Charts: I am trying to keep this thread alive, as a kind of 'one-stop shop' with essays on how to use BC, Technical Analysis, Trading Tips and Ideas, and links to a wealth of other relevant information. Contributions from allcomers welcome! ESSAY # 6.1 Emotion Free Trading Book: Some excellent articles by Ruth Barrons Roosevelt: Article: "Master the Four Fears of Trading" Article: "Psychology & Behavioral Finance" Article: "Essential Characteristics of the Successful Trader" by Joseph Stowell http://www.bondtrades.com/essent/essent1.html Trading Psychology articles section from Action Forex: http://www.actionforex.com/contentcategory/trading_psychology_articles/ Happy reading
This contains links to a variety of other web-based resources.
Useful reading for newbie traders: A wealth of reference links here:
There is also a link there to a free book, which encapsulates most of the above:
RULES FOR SUCCESSFUL TRADING Dennis Gartman's rules for successful trading
NOTES: Bullish divergence is the converse, i.e. when price falls to a lower trough than the previous one, but the oscillator forms a higher trough. This can sometimes point to weakening downward momentum, and herald the end of the falling trend.
Mr Hull's own web page can be found at:
The BULLETIN hyperlink there gives some very good general purpose articles on trading systems. There is also a FORUM, and a list of around 20 SYSTEMS (i.e. entries and exits) that have been tested across several years of historical data, and may be purchased.
Some good articles here: Discussion boards:
This (Australian?) gentleman appears rather skeptical of TA, so some of the articles make for a good cautionary read, for any newbie thinking that making money from trading is going to be easy. Of particular interest, see the following: Murphy's article, and his 20 trading rules: HERE IT IS! A definitive (if dated) study as to how many futures traders actually succeed, and why those who fail, do so: The index to the trading FAQ is here:
List of technical indicators (on left sidebar) List of entry signals/systems pre-programmed into Stockworm (good if you want some ideas to get started with designing your own system - see left sidebar):
Click on the appropriate topic ('Technical Analysis' is a good start), to get a raft of hyperlinks and article references to that topic. If you have a technical term that you don't understand, you can be sure to find a huge number of explanatory links here.
In particular, these articles are particularly relevant to BC's AIO..... ....because they discuss the pros and cons of curve fitting, which is essentially what AIO does. This is an EXTREMELY IMPORTANT read for those who intend to use AIO!
Homepage: Setups page: Articles page:
Free Trading Articles: (see the Navigation Menu hyperlinks on the left of the screen for more goodies)
* Chart of the Week
MATERIAL ON ELLIOTT WAVE THEORY
[This message has been edited by david_louisson (edited 12-12-2006).] IP: Logged |
|
david_louisson Member Posts: 303 |
posted 11-28-2004 03:34 AM
ESSAY # 7.1 ===== INTERPRETING THE OUTPUT GENERATED BY BEST-CHARTS ===== BC’s functions generate a number of viewable results, which are also stored as HTML and TXT files, in the C:\M-C\ folder. Here are some notes on how to interpret them. Remember that HTML files can be loaded directly into Excel for further columnar and other analysis. -------------------------------------- Output columns and what they mean: 1. Date: the date on which the pattern occurred. 2. Pattern: the name of the pattern. For information on what the pattern means, go to http://www.litwick.com/glossary.html 3. Trend: whether the current trend, as determined by the pattern, is Bullish (upward) or Bearish (downward). 4. Direction: whether the pattern is heralding a CONTINUATION, or a REVERSAL, in the current trend. 5. Credibility: how reliable the pattern’s forecast is, based on a statistical analysis of a large sample of these patterns. Possible values are Low, Moderate or High. Notes: a) It is important that the candle pattern is backed by signals and trends provided by the other BC indicators. b) For a more complete understanding of what a pattern means, it is important to read the accompanying explanation in: http://www.litwick.com/glossary.html
There are two tables output: I. Table of trades based on entries and exits calculated for the SELECTED INDICATOR and its parameters (Set1, Set2, etc or AIO), based on the BT options selected (number of quotes, which of long and short positions are to be considered, etc). II. Table showing the comparative gains for ALL INDICATORS, based on the BT options selected. Output columns and what they mean: Table I 1. No: the sequential trade number. 2. Trading: either Long (expectation of rising price) or Short (expectation of falling price). Which of these appear depends on whether you have selected Long, Short or Both in the BT parameters input. 3. Entry Date: the date the position (trade) was opened, which occurs when the buy (green O for a long position), or sell (red X for a short position) signal. Note that if you are trading both Long and Short positions, then the exit (closing) date of one trade will be the entry (opening) date of the trade immediately following. The entry date is determined according to your input, i.e. whether you have chosen the ‘signal day’ or the ‘day following the signal’ in the BT options. 4. Entry Price: the price when the position is opened, determined by whether you have selected ‘average price on signal day’, ‘closing price on signal day’ or ‘opening price the day following the signal’. Note that if you have selected both Long and Short positions in the BT parameters, then the exit price from the preceding trade becomes the entry price for the current one. 5. Exit Date: the date the position (trade) was closed. This occurs ONLY when the opposing signal is generated, i.e. a ‘sell’ (red X on the chart) to exit a Long position, or a ‘buy’ (green O on the chart) to exit a Short position. The actual date can be either the day the signal was generated, or the day immediately following, depending on whether you have chosen the ‘signal day’ or the ‘day following the signal’ in the BT options. See also notes below about lack of stop losses. 6. Exit Price: the price when the position is closed, determined by whether you have selected ‘average price on signal day’, ‘closing price on signal day’ or ‘opening price the day following the signal’. 7. Gain%: this is the gross percentage return on the trade. 8. Total Gain%: This is the cumulative gain across all trades to date. Results are COMPOUNDED rather than simply being added, i.e. Result(n) = Result(n–1) x ( 1 + Gain%(n) ) Table II 1. Indicator: the name of the indicator being compared. 2. Total Gain%: calculated, for each indicator, as shown in point 8 in the previous section. 3. Number of trades: under certain conditions, this can be a little misleading, as the table considers each entry and exit as being one trade. Hence the values shown in the column are actually double the number of trades executed. 4. Gain% / Trade: calculated as = Total Gain% / Number of Trades (i.e. column 2 divided by column 3). Notes: The purpose of back-testing is to allow you to see the profits (or losses) given by the applying different parameters (e.g. AIO, unoptimized) against different indicators, giving a head-to-head comparison by indicator type. In evaluating the results, the following should be kept in mind – a) No stop losses are being considered. Many trades resulting in losses could have been stopped out earlier. If there have been trades showing substantial losses, this will have the effect of UNDERSTATING the overall Total Gain%. b) No transaction costs (brokerage, spread, slippage, borrowing costs) are taken into consideration, which will have the effect of OVERSTATING the overall Total Gain%. The more trades that are generated by the indicator signals, the greater the effect, since costs are normally incurred on each transaction. c) Remember that (i) if AIO has been used, profits shown are MAXIMIZED rather than REALISTIC gains (see section of post above entitled ‘Benefits and Caveats of Dynamic Optimization’ for a full explanation), and (ii) that profits are being overstated or understated across the board for all indicators, hence this remains a valid basis for a head-to-head comparison of the potential performance of each indicator. d) In my view, ‘Gain% / Trade’ is a more valid basis for comparison than ‘Total Gain%’. This is because transaction costs are not being considered (see note (b)). e) The parameters for each indicator that were used to perform the TA are carried over into the back-test. Unless you have run an AIO prior to the BT, it is most likely that performance is being calculated using the default parameters (Parameter Set 1).
There are two tables output: I. Current forecast for each stock: A summary of the current trends (bullish or bearish), probabilities, candle patterns and signals based on the back-testing performed for each stock in the portfolio, ranking the stocks from ‘highest upward trending probability’ at the top of the table to ‘highest downward trending probability’ at the bottom. II. Profitability back-testing by indicator: A summary of the gain or loss generated by each indicator, according to the back-testing parameters used, across each stock in the portfolio. Output columns and what they mean: Table I 1. Symbol: the ticker code. 2. Quote: latest price quoted by Yahoo (or the historical/EOD file) for the stock. 3. Date: date of the latest quote (or time, if the date = today). 4. Probability Advance/Decline: this probability is based in same way on the number of bullish and bearish signals (see point 6), although I am unsure of the exact calculation. The sum of the advance (price will rise) and decline (price will fall) values should always be 1. The table is sorted by default so that the stock with the highest probability of advance is listed first, and that with the highest probability of decline last. 5. Signal number Buy/Sell: the number of buy (green O) and sell (red X) signals that occurred today on the charts of the stock in question. 6. Trend Bullish/Bearish: when a buy signal (green O) is generated, the indicator is considered bullish (price beginning to rise). It remains ‘bullish’ until a sell signal (red X) is generated, at which point it is considered bearish (price beginning to fall), and so on. This column gives the total number of indicators that are bullish or bearish for the stock in question. Generally, this will total 15 or 16, as there are 16 indicators plotted for each stock (see point 8). 7. Candle: whether a recognizably bullish or bearish candle pattern has occurred today (most likely result is neither, resulting in a blank entry). BC tests for around 60 candle patterns. You can run ‘Stocks > > Display candlestick chart signal list’ for a stock to see a candle pattern history for a stock. To understand the meanings of the candle patterns used by BC, see http://www.litwick.com/glossary.html 8. BBI, MA, ITA, etc: This tells whether a specific indicator is in a bullish or bearish phase of the cycle (see point 6 above). A ‘1’ (or ‘2’) indicates that the number of buy (if Bullish) or sell (if Bearish) signals that occurred today, while the symbols ‘ob’ and ‘os’ are abbreviations for ‘overbought’ and ‘oversold’. Overbought means that the stock is at a prolonged or extreme high, and that a reversal downward (sell signal) will eventually occur, while oversold means that the stock is at a prolonged or extreme low, and that a reversal upward (buy signal) will eventually occur. Table II The parameters selected for the back-test are listed in blue typeface at the top of the table. 1. Symbol: the ticker code. 2. Parameters: the parameter set used for the back-testing, e.g. AIO or unoptimized (Parameter Set 1). 3. BBI, MA, ITA, etc: the Total Gain% or Total Loss% (losses have a preceding minus sign) that occurred during the back testing for the indicator (column in the table), for the stock in question (row in the table). You can use this data to determine which indicators are currently working best with each stock in the portfolio, or which indicators are delivering the most profitable overall result. Analyzing a portfolio goes through the same calculation process as back-testing a single stock, except of course that the process is run against each stock in the portfolio.
This produces exactly the same two tables as ‘TAsummary.html’ (see previous section), except that instead of there being one row in the table for each stock in the portfolio, there is one row for each different set of back-tests of the stock. Each set involves the use of a different number of days (‘quotes’). [Note: multi-AIO allows back-testing of each indicator across differing numbers of days, allowing optimization not only by indicator(s) but also by number of days]. The ‘number of days’ is shown in the leftmost column of both tables.
This is a comma separated file that is best loaded into Excel, and re-formatted using commas as delimiters, for further analysis. It gives insight into the workings as to how the buy and sell signals are calculated for the different indicators during a TA. The leftmost two columns (A and B in the spreadsheet) contain much the same data as in Table I of TAsummary.html (see relevant section above), except that the columns have been transposed into a vertical arrangement, thus – Symbol ^FTSE The columns from C rightwards give actual values for each indicator and/or component, for the date given in column D, thus – Column C – No. : a sequential number (probably the array element used in BC). See a previous topic ‘How the Indicators and Signals Work’ for more information. ===== END OF POST ===== IP: Logged |
|
david_louisson Member Posts: 303 |
posted 11-28-2004 04:38 PM
ESSAY # 8.1 ===== IMPORTANCE OF STOP LOSSES AND POSITION SIZING ===== Because it is impossible to predict the direction of market prices with absolute certainty, not every trade will result in a gain, and no system of calculating entries and exits is perfect. Even the very best systems encounter periods where losses outweigh the gains. Experienced traders follow the maxim of ‘letting profits run and cutting losses short’. In other words, if a trade moves in the wrong (losing) direction, they will exit quickly, and look for opportunities elsewhere. The market is always offering a plethora of opportunities; smart selection is important. Pride and/or false hope are a trader’s enemies. Inexperienced traders may stay in failed positions too long because they don’t want to admit to themselves that they have made a bad choice, or because they feel that ‘the price can’t get any lower, it must turn around any day now, so I’ll hang in there’. Setting mechanical STOP LOSSES can help avoid this kind of mistake – if the price falls to a previously determined level, you make a disciplined exit – no hesitation, no excuses, no regrets. Stop losses are one means of protecting your capital pool. Correct POSITION SIZING is another. What this means is that the trader is risking only a small percentage of his capital on each trade. If a trade goes wrong, he exits at his stop-loss point, safe in the knowledge that he has only lost a pre-determined fraction of his capital. Put simply, you will be able to continue trading, so long as you are able to safeguard your capital. If you expose too much of your capital to risk, it can take only a few big losses to wipe out your entire trading account. The importance of exercising PATIENCE and DISCIPLINE can not be stressed highly enough. Trading is about understanding, and balancing, RETURN and RISK. This essay discusses the use of stop losses and position sizing, with a view to optimizing this. There are two sets of important parameters associated with trade win/loss math. The first is the ratio of wins to losses, and the second is the average size of winners relative to the average size of losers. In discussing wins and losses, I am referring to the NET result of a trade after the position has been closed and any transaction costs (e.g. brokerage, spread, slippage, borrowing costs) have been deducted. Here are two examples: Trader A makes wins of 4%, 3%, 7%, 5%, 3%, 2%, 10%, 1% and losses of 5%, 11% and 20%. He has made 7 wins and 3 losses, i.e. a 70% win rate. His average win is 35 / 7 = 5%, and his average loss is 36 / 3 = 12%. One way of calculating his net position is = 35 – 36 = -1%, i.e. despite the 70% win rate, he has made an overall net loss. Trader B makes losses of 2%, 2%, 2%, 2%, 2%, 2%, 2%, 2%, and a wins of 10% and 20%. He has made 2 wins and 8 losses, i.e. a 20% win rate. His average win is 30 / 2 = 15%, and his average loss is 16 / 8 = 2% (this is because he is setting a stop loss to exit any trade immediately 2% has been lost). Calculating his net position in the same way, overall he is 30 – 16 = 14% ahead. These examples are designed to illustrate the importance of (1) using stop losses, and (2) that the average win to loss size is every bit as important (if not more so) than the winning percentage. In other words, it is possible to have a profitable trading system that selects winners less than 50% of the time, provided that the few winning trades are returning high enough profit to offset the high number of smaller losses. Of course, the objective is to devise and operate a system that delivers BOTH a > 50% win rate, AND an average win size > average loss size. Another one of a trader’s enemies is ASYMMETRIC LEVERAGE. To explain this, if one makes a 20% loss, it takes a 25% gain to bring the account size back to break-even point. Do the numbers to satisfy yourself that this is correct. If one starts with $100 in a trading account, and encounters a 20% loss, there is now only $80 to trade with. To break even, one must generate a $20 win, but with only the $80 to play with, hence a 25% win is required. It is not difficult to see how a succession of losses could cause spiraling drawdown, and eventual wipe-out of the account.
There are many ways of calculating stop loss points, but a simple and effective one is to set your stop just below a point of previous SUPPORT. For an explanation of support and resistance, see http://www.incrediblecharts.com/technical/support_resistance.htm Why? Because the market is driven by crowd behavior, and many players will tend to start ‘buying in’ once a stock has rebounded off a previous low, using the logic that ‘if buyers were interested at $5.00 before, they will be interested as soon as the price falls to this point once again’. The added buying pressure at this point should start to drive prices upward once again. However, if the prices tend keep falling after this ‘support level’ has been penetrated, then it is reasonable to assume that they will continue to fall, and consequently the position can only get worse. Hence it is logical to pre-set your stop loss at this point.
Using the same kind of reasoning in reverse, a winning trade is likely to encounter RESISTANCE when prices start to reach a previous high. Hence, players will likely start to sell a stock at this point, so you can set an imaginary target win there. I say ‘imaginary’, because of course if your trade does ultimately reach this price, you will of course continue to take profit gratefully until such time as there is evidence of a reversal ACTUALLY occurring. However, the imaginary target (resistance point), and the stop loss (support point), can help you choose between two or more possible candidate trades. Consider the following example: Trade A: price currently $2.00, most recent support at $1.75, most recent resistance at $3.25 In other words, trade A is offering a potential ($3.25 - $2.00) / ($2.00) x 100% = 62.5% gain, for a ($2.00 - $1.75) / ($2.00) x 100% = 12.5% loss. Given that 62.5% / 12.5% = 5, this means that the upside potential (‘return’) is 5 times the downside potential (‘risk’), i.e. there is a 5:1 ratio for success. We say that the trade has an ‘R-VALUE’ of 5. Trade B is offering a potential ($5.20 - $5.00) / ($5.00) x 100% = 4% gain, as opposed to a ($5.00 - $4.75) / ($5.00) x 100% = 5% loss. Hence the R-value for this trade is 4% / 5% = 0.8. Hence, trade A offers significantly greater potential than trade B. If both were to give a buy signal, A would likely be the better candidate. Of course, a trade may not necessarily reach the forecast target, but by applying this method you are allowing a greater margin for error, and thereby increasing the probabilities in your favor. In reality, time is also a factor. A candidate with an R-value of 5, but is taking 6 months to cycle between resistance and support points (in the timeframe you are trading) is arguably inferior to a trade that has an R-value of 2, but whose cycle time is 2 weeks. The second trade is a more volatile stock, and is therefore likely to deliver the profit MORE QUICKLY.
This section deals with how much of your capital you should place at risk on a single trade, i.e. the ‘size’ of the position. As a general rule of thumb, no more than 2% should be risked on any given trade. If the R-value is high, and/or you are trading in an upward longer-term trend, and/or the sector and/or market as a whole is rising, there may be a case for increasing this a little. Diversification is a further alternative for increasing return disproportionately to risk (see next section). Let me restate this clearly: increased position size = increased potential return = increased risk. Return and risk are proportional to each other. Mathematically, the optimal system is one that maximizes return without risking irretrievable drawdown. However, in reality there are other considerations: firstly, price movements and market probabilities are, at best, only mathematically approximate; secondly, high drawdown, and wild fluctuations in one’s account balance, make for uneven cash flow, and an uneasy night’s sleep! The aim of conservative position sizing, limiting risk to 2% per trade, is to protect your capital pool, and keep stress at a manageable level. Let us assume that your trading account balance (i.e. capital pool) currently stands at $10,000, and that you are considering Trade A in the example above. Placing 2% of your capital $10,000 at risk means that your downside target is 2% x $10,000 = $200 maximum loss if your stop loss is hit. With support at $1.75, the price must fall ($2.00 - $1.75) or $0.25 for this to occur. Hence you can buy $200 / $0.25 = 800 units of the stock. Convince yourself of this by working the calculation in reverse: 800 units x $0.25 loss per unit is a loss of $200, which is 2% of your total $10,000. Ideally, you should also factor any transaction costs into this calculation, so that the amount placed at risk is your REAL loss, should the stop loss point later be reached. On the upside, if the price does reach the forecast target, your gain will be 800 x ($3.25 - $2.00) = $1,000. To check the calculation, the gain would $1,000 / $10,000 = 10% of the account. Whichever way you do the math, it stands to reason that $1,000 gain / $200 loss = 10% gain / 2% loss = 5:1 ratio, i.e. the R-value. A potential 10% gain on one’s total capital from a single trade is not too bad! This illustrates that significant gains can be made even though only 2% is being placed at risk. Note that: 1. It is the POSITION SIZE that determines the overall gain relative to your account size, not the percentage movement in the stock price. 2. Return and risk are linearly proportional, i.e. in the above example, if you had bought double the number of shares (1,600), then the potential gain doubles, but so does the potential loss. 3. By operating in this fashion, or something similar, you maintain a degree of control over your account, and what your maximum loss per trade can be. You CANNOT control the price movements in the market, but you can control when you enter and exit, and – most importantly – the amount you are placing at risk.
It is possible to increase the number of gains further, by having multiple 2%-risk positions open simultaneously, but with the following caution: stocks, and markets tend to move in harmony with each other to some extent, so the trades may not be completely independent. In other words, if the price of one stock falls, then the market, or sector, may also be falling in unison. So effectively more than 2% is being placed at risk. A good compromise might be to have, for example, 5 positions open at 1% risk each, at any one time. This puts a total of 5% of your capital pool at risk, but it is diversified across 5 trades. If the stocks involved are in different sectors (e.g. technology, real estate, retailers, etc) and different markets (e.g. US, UK, Japan) then to some extent, the effect of (undesirable) correlation is decreased. Throughput is related to time frame. A day trader will likely put through several trades during the course of a day, giving frequent opportunities for small gains (or losses), while a longer term investor will make bigger gains (or losses), but less often. The former is, in effect more diversified (‘safety in numbers’), while the latter may experience poor cash flow for a prolonged period if two or three consecutive losses are encountered. Offsetting this, greater throughput means more intensive position management, which can be laborious (although computers can help automate this). Much of it comes down to a lifestyle decision.
I believe that there is a case for compounding gains (and losses!), provided you are confident that your system is delivering POSITIVE EXPECTANCY (total gains exceed total losses, across a period long enough to be statistically significant). Whether you are putting $200 of a $10,000 account at risk, or $20,000 of a $1,000,000 account, the risk is still the same: 2%. However, it is important that one trades at a level that one feels comfortable with. Anxiety can affect one’s ability to trade mechanically to a pre-determined plan. Note that BC's back-testing calculates Total Gain% by compounding gains and losses.
The following are tangible transaction costs that must be overcome before one makes a profit – Brokerage: commission paid to a broker in order to buy or sell stock. Spread: if the stock is listed at $9.10 / $9.30, there is a $0.20 spread. Because you must buy at the high price, and sell at the low price, the spread is a further cost. Slippage: the inability to buy or sell stock at the listed price, due to lack of liquidity. For example, stock ABC is currently listed at $2.00, but you want to buy 1,000 units. There are only 500 units available for purchase at $2.00, and 500 units at $2.02. So your effective (average) buy price is $2.01, not $2.00. The reverse effect can occur when you are selling. Borrowing costs: if you are trading leveraged instruments (options, futures, CFDs, etc), or stocks ‘on margin’, the broker or financial bookmaker will likely charge interest for the funds borrowed. For example, to buy $100,000 worth of stock ABC, the bookmaker may require that you only have $20,000 (i.e. you are buying on a 20% margin) in your trading account, but you will need to pay daily interest on the $80,000 at the specified rate. Here are some not-so-obvious costs that must also be considered – Trend establishment / reversal cost: you can not hope to buy at the bottom of the curve, because you need evidence that the former downward trend has actually reversed, i.e. a new upward trend is beginning. Hence you must sacrifice some potential profit in terms of the difference between the absolute low, and the later entry point. The same principle applies when you are closing a position. You want to ‘let profits run’, so the aim is to ride the trend until there is definite evidence of a reversal. Hence you can not hope to sell at a proven high point, so some profit is lost in the process.
Some derivative instruments (options, futures, CFDs, spread bets etc) allow the trader to take positions in such a way as to profit in a falling market. This is commonly called short selling, and is effectively a bet that the stock price will fall rather than rise. In other words, if the stock falls, you make a profit, if it rises, you make a loss. All of the above – calculation of stop loss point, R-value, and position size – applies exactly the same, but in reverse.
Much of the psychology in trading involves how you deal with the inevitable losses. I like to think of trading as running a retail business. I try to dismiss losses philosophically, paralleling them as necessary day-to-day overheads like purchasing of stock, and wins as making sales to customers. As long as the sales exceed the costs, the business will profit in the long term. It is important to realize that when a stop loss is reached, although your account has made a loss, you have made a correct trading decision. THAT IS SUCCESS, NOT FAILURE. Given that your forecasting system is capable of picking sufficient winners with a high enough R-value to overcome costs, you can rest confident in the knowledge that, provided you manage your position sizes conservatively and correctly, the probabilities are in your favor, and like a casino, the wins will ultimately outweigh the losses in the longer term. See also a previous post (essay 6.1) which has links to some excellent reading on trading psychology.
Possibly one of the bigger mistakes made by inexperienced traders is that they move their protective stop too hurriedly to the break-even point. As an example, suppose a trader enters at $10, with the protective stop at $9.70. Then when costs are covered as the price hits $10.10, he immediately moves his stop to $10.00, so that (slippage apart) he can not lose. In other words, price has moved only 10c, but he has advanced his stop by 30c. IMHO, this is an error. If there was a valid reason to trail the stop by 30c at the entry point, then it is unlikely that volatility has changed sufficiently to justify suddenly tightening the stop to just 10c below the current price. A better option would be to move the stop to $9.80, or (when appropriate) to just below the next swing low. Whatever the rationale, the reason should be a technical one, as opposed to simply averting a losing trade. As a wise man once said "it's not how many trades you win, but how much you win on each trade". Or, put another way: it is neither more nor less valuable to lock out 10c worth of loss than it is to lock in 10c worth of profit, as each ultimately has an identical effect on eventual bottom line. But somehow, fear of loss gets in the way, and we do not see this clearly. Unduly tightening a stop increases the risk of being stopped out prematurely, with the "double whammy" effect that not only was a winning opportunity squandered, but a loss is incurred in its place. It violates the maxim of letting profits run. Of course one can always re-enter the position, but that means incurring a second set of costs.
===== ADDITIONAL MATERIAL ON POSITION SIZING ===== As discussed above, ultimately it is how you size your positions, and manage both capital and risk, that will determine your overall profit. It is possible, through good TA and some luck, to string a few winning trades together, but if you want to be successful year-in-and-year-out, then you will need to have both a discplined system of entries and exits, and also devote some study to how you manage risk, your trades, and your capital. The following site takes some of these management aspects further, providing some very useful information on position sizing, Monte Carlo analysis, trade dependency, and significance testing. A very worthwhile read, in my view. http://www.adaptrade.com/Articles/ Note that the position sizing formula that I described in the previous essay (8.1) is documented here as the 'fixed fractional' method. I have read the articles (which are free), but haven’t used the software (which is not).
[This message has been edited by david_louisson (edited 03-05-2006).] IP: Logged |
|
david_louisson Member Posts: 303 |
posted 11-30-2004 04:43 AM
ESSAY # 9.1 ===== MULTI-AIO ===== Multi-AIO is a new facility in BC revision 4.33. When you click the BT button, one of your inputs is the number of quotes (price bars) over which the back-testing is to be run. If you set this too high, any optimization will be giving the oldest data in the window equal priority with the most recent. If you set this too low, then you will not be analyzing enough price bars to attain statistical significance. You can use Multi-AIO to try up to 8 different entries for this parameter, for the current stock. Click ‘Stocks > Multi-AIO’ and type in your 8 entries. The process takes some time, while 8 different AIOs are run. Multi-AIO generates an HTML-based result whose second table shows the profit(/loss) for each indicator when back-tested across each of the 8 entries. You can scan the results, or load the generated file (C:\M-C\MAIO.html) directly into Excel for analysis. Summing or averaging each row (applying weights however you wish) will tell you which of these values should be entered into the ‘Number of quotes for BT and AIO’ parameter (set when you click the BT toolbar button). You can then run an AIO in the normal manner, specifying this value. Remember the specific purpose of each function: * Multi-AIO – to find the optimum number of quotes (price bars) over which to run the back-tests * BT – to provide head-to-head profitability comparison of each indicator * AIO – to optimize the parameters for each indicator, by way of further profitability back-testing * TA – to show the buy/sell signals for each indicator, and Bullish/Bearish counts, based on the current (optimized or unoptimized) parameters
[This message has been edited by david_louisson (edited 01-08-2005).] IP: Logged |
|
david_louisson Member Posts: 303 |
posted 12-04-2004 06:50 PM
ESSAY # 10.1 ===== HOW CRITICAL IS CORRECT PARAMETER CALIBRATION IN TERMS OF OVERALL SUCCESS? ===== Apart from visually displaying OHLCV, indicators, trendlines and other studies, and symbols, the best that any charting tool can do is use past history to generate buy and sell signals. BC goes further than many in that it also gives back-testing and AIO facilities. It is very reasonably priced compared to the likes of MetaStock, AmiBroker, TradeStation, etc. To answer the question ‘do you need to change the default parameters?’, I will use RSI as an example. For better or worse, most traders that I know simply use the default settings supplied by their charting package, unquestioningly. For RSI that value is 14, and I notice that it is also the ‘factory’ default in BC. Because it has a highly automated AIO function, BC offers incentive to move from default to optimized parameters. Many traders use oscillators like RSI, to highlight oversold/overbought situations, and also divergences. You can of course use BC’s RSI plots to (manually) look for these, just as you would with any other package. But BC also uses RSI to generate buy/sell signals when RSI crosses a MA of itself, which is a novel idea. I have used RSI as an example, but the other indicators work similarly. I will compare some of BC’s factory defaults with those of MetaStock (where appropriate): BB: both BC and MS default to a MA of 20 (with 2 standard deviations) Of course, it is easy enough to change the default settings in either product. There is no ‘holy grail’ set of parameters, because different stocks move in different cycles, and these cycles are constantly changing. That is what AIO is attempting to capture, by using the technique of profitability back-testing. Back-testing that is run over too small a statistical sample can occasionally be meaningless. For example, across a short time period there might be one short, sharp price movement that accounts for a disproportionately huge gain. A certain parameter setting might catch this price movement early enough, while another might generate the buy signal a day or two too late, causing the profit to be missed. That is part of the fickle nature of TA. You can set parameters to generate either earlier or later entry / exit signals. Earlier ones (before a reversal has firmly established itself) will catch greater profits when they do work, but will also fail more frequently because they are more likely to pick up ‘noise’ (i.e. undesirable minor fluctuations) along the prevailing trend. Later entries and exits will tend to generate lower profit trades, but will be more consistently successful. Reaching a compromise is a finely judged balance. In any trade, you will make money if and only if the price trends far enough between the entry and exit to cover your costs. If the trend peters out too quickly, even high probability entries and exits can result in losses. That is not the fault of the buy / sell signals, the indicators or parameters used, or the charting package. You can use the likes of trend lines, previous support/resistance, or Fibonacci retracements, to try to guess how far prices will move, but the key word is ‘guess’. News can cause unexpected or chaotic short term price movement that can undo even the most conscientiously applied TA. The best you can do is use a combination of techniques to move the probabilities as far as possible in your favor, and then size your positions conservatively enough to ride out any temporary run of losses. Finally, buy / sell signals are but the tip of the trading iceberg. Whether you ultimately profit is determined by a number of other factors, that are not directly addressed by many charting packages, e.g. * How do you manage risk, and exit losing positions (e.g. stop losses)? In terms of overall profit, virtually all of the above are every bit as important as how indicators used, or parameter settings, are affecting your entries and exits. ===== END OF POST ===== [This message has been edited by david_louisson (edited 01-08-2005).] IP: Logged |
|
david_louisson Member Posts: 303 |
posted 12-11-2004 04:53 AM
ESSAY # 11.1 ===== UNDERSTAND THE INDICATORS USED IN BC ===== This essay attempts to show how the indicators used in BC are calculated, what they measure, and how to interpret them. Rather than try to explain all of this myself, I have simply included a number of URL references, for each indicator. Note that BC uses many of the indicators in an unconventional way, in that it uses crossovers of an MA of the indicator with the indicator itself, to generate a buy or sell signal. Examples are RSI, SO, CCI, MFI and OBV. The material that follows should give you additional information on how the indicators are ‘normally’ used by experienced traders, e.g. to measure divergence, volatility, momentum, etc, which should improve your knowledge, and add to your trading arsenal. Among these are some websites that offer comprehensive TA primers, namely Note to webmasters: I have nothing to gain financially by including references to your sites. However, hopefully their inclusion might alert readers to any products and services you may be offering. If you are unhappy about this, post a suitable reply, and I will remove all references to your site. David Louisson
TECHNICAL INDICATORS MOVING AVERAGES INTELLIGENT TECHNICAL ANALYSIS (ITA) BOLLINGER BANDS VOLUME MA ENVELOPES MOVING AVERAGE CONVERGENCE/DIVERGENCE (MACD) CANDLESTICKS STOCHASTIC OSCILLATOR (SO) RELATIVE STRENGTH INDICATOR (RSI) COMMODITY CHANNEL INDEX (CCI) DIRECTIONAL MOVEMENT INDEX (DMI) BEST-CHARTS INDEX (BCI) PRICE RATE-OF-CHANGE (ROC) ELDER FORCE INDEX (EFI) WILLIAMS %R (WMS) PERCENTAGE PRICE OSCILLATOR (PPO) MONEY FLOW INDEX (MFI) ON BALANCE VOLUME (OBV) PRICE & VOLUME TREND (PVT) CHART PATTERNS GAPS
[This message has been edited by david_louisson (edited 01-21-2006).] IP: Logged |
|
Admin Administrator Posts: 1091 |
posted 12-11-2004 09:27 AM
Dear David; Thank you for your good articles! IP: Logged |
|
david_louisson Member Posts: 303 |
posted 12-12-2004 06:11 PM
ESSAY # 12.1 ===== NAVIGATING YAHOO FINANCE ===== The following illustrates the general URL syntax for navigating stocks in Yahoo Finance. To obtain the item shown at left, enter the URL on the right: Quote summary for a stock: http://finance.yahoo.com/q?s=[stock-symbol] Where [stock-symbol] is the symbol involved (without the angle brackets). If you don’t enter this, Yahoo will prompt you for it. For a list of major world indices, see http://finance.yahoo.com/m2 For a list of exchange suffixes, see http://finance.yahoo.com/exchanges Note that, when displaying a stock, there are also other hyperlink options available, e.g. to get Microsoft’s company profile, either load Microsoft using any of the above links, then click the hyperlink; or simply proceed directly by http://finance.yahoo.com/q/pr?s=MSFT The additional charting options can be likewise obtained via the hyperlinks, or by including them after the stock symbol, and separating them with & symbols, thus: For example: http://finance.yahoo.com/q/ta?s=QQQQ&t=1d&l=off&z=m&q=b&p=v,b&a=m26-12-9,ss,w14,f14&c= Some other useful URLs in Yahoo Finance: ===== END OF POST ===== [This message has been edited by david_louisson (edited 12-12-2004).] IP: Logged |
|
alvin_chiu unregistered |
posted 12-16-2004 09:33 PM
Dear David, This article is unbeatable! I must give you a credit. I just archive it and read over and over. Thank you for your information, it is structurize and well-organized. Even I had quit watching the market some times ago due to work affairs, I am coming back and learn Best Regards, IP: Logged |
|
david_louisson Member Posts: 303 |
posted 01-10-2005 08:51 PM
Thanks for the feedback. Happy New Year to everybody. David IP: Logged |
|
vijayala unregistered |
posted 01-11-2005 08:46 AM
Dear David Highly informative and one-shop stop article. I was searching for candlesticks . Your references are a great help to me. Please continue to post your trading strategies with optimum parameter settings of BC. BC works pretty fine for going long. Day trading with BC requires more precise displays of the charts. vijayala IP: Logged |
|
david_louisson Member Posts: 303 |
posted 01-11-2005 04:06 PM
Essay # 13.1 ===== DOES TECHNICAL ANALYSIS REALLY WORK? ===== NOTE: The reading that I have done leads me to believe that very few 'serious' traders enjoy consistent, long-term success at this occupation, and that this percentage could be as low as 5%.
Many professional financial advisers and commentators (at least in my part of the world) have been taught that short term price movements are COMPLETELY RANDOM, and that the only way to profit consistently in the stock market is through long term holdings. In other words, they think Technical Analysis is a complete WASTE OF TIME. When confronted by evidence that some traders do profit consistently, their arguments are – 1. That for every dollar lost, there has to be a dollar won somewhere else. If we were to draw a normal distribution curve (remember Statistics 101?), then given the assumption that short term price movements are completely random, the fact that 5% of traders are winning long-term (thus far) remains perfectly within the realms of 'expected' probability. [Analogy: if 100 people were to each toss a coin (i.e. an event with a completely random outcome) 100 times, it is more likely than not that a few people (e.g. 5 out of the 100) would toss an ‘abnormally’ large number of heads]. In other words, they are saying that the 5% have merely made more lucky guesses than unlucky ones (and the implied assumption is that, given enough time – maybe several years – their luck will eventually change). 2. Given the transaction costs, trading is actually a negative sum game. Over time, these costs become significant, and this accounts for the other 95% who blow their bankroll, and fall by the wayside. 3. Yes, trends do appear to be obvious on price charts, but this is based upon the view of hindsight. One can always position and calibrate trendlines, indicators, channels etc to show only the profitable trades. The trends – by their very definition – are overtly conspicuous, the ‘failures’ sandwiched in between them are not. 4. Because of the randomness, market cycles and rhythms are constantly, but gradually, changing. Systems that work today may not work in several weeks, or months, time. This is evidenced by those (and there are apparently many) who are buoyed by a season of profit, complacently believing that they have ‘cracked’ the market, which is followed by an unexpected and devastating sequence of losses. 5. The fact that significantly more people make money out of TA by selling books, systems, software, tuition, etc, rather than by actually trading. This begs the question: if their knowledge and systems are as good as they claim, why aren't they simply trading their way exponentially toward a fortune? Moreover, much of the material available offers generalized principles and guidelines, as opposed to a single, definitive trading methodology that is guaranteed to deliver consistent success. 6. Supporters of FUNDAMENTAL analysis argue that there is some tangible, longer term, underlying basis for a rise or fall in a stock’s price, because the current price can be, in ‘real’ financial terms, be shown to be over or under valued. 7. Technical analysts are apparently divided as to whether trading is an ‘art’ or a ‘science’. Moreover, the vast number of different systems around is testimony to a lack of consistency in approach; and this lack of consistency is a reflection of uncertainty, which is ultimately rooted in the randomness of the market. 8. Wishful thinking is rooted deeply into the human psyche, and technical analysts are continually tantalized by the expectation of exponential gains, and the hope of an eventual life on ‘easy street’. No-one has yet provided conclusive scientific proof that TA works, and science demands that the burden of proof must lie with the proponent of a theory.
Those supporting Technical Analysis argue that trading is demanding both intellectually and emotionally. The 95% that end up on the scrap-heap do so for any or all of the following reasons – 1. They have entry/exit systems that fail to deliver long-term positive expectancy, and/or fail to use proven systems that have been tested properly across all the different types of market (rising, falling, trending, ranging, volatile, stagnant, etc). 2. They fail to manage risk properly, i.e. do not employ correctly calculated stop losses. Put another way, they exit losing trades too slowly (instead of ‘cutting losses short’), and winning trades too quickly (instead of ‘letting profits run’). 3. They fail to size their positions conservatively enough, so that it only takes a short series of losses to wipe out their account. Viewed another way, they are under-capitalized for the level of trading that they are attempting. 4. They lack the discipline, patience and nerve to trade mechanically to a plan. In other words, they are swayed by anxiety, greed, pride, impatience, complacency, discouragement, and also by hunches, outside ‘buzz’, etc. 5. They fail to follow simple maxims like trading in the direction of the long term trend. Alas, occasionally one does profit when one breaks such rules, providing incentive to continue flouting them – with the inevitable end result. 6. For ignoring the impact of subtle but significant psychological aspects, e.g. they fail to view trading as a probabilities game, and get too emotionally involved with each individual trade; or perhaps they set unrealistic profit targets, and thereby overtrade; or they focus obsessively on profit, rather than simply following the correct trading rules. Potentially, there are a host of other psychological reasons, depending on the personality make-up of each individual. 7. TA proponents agree that trading is a (close to) zero sum game, and the fact that the 5% who are winning are doing so at the expense of the 95% who are losing is simply the result of the highly skilled (i.e. those with superior information, tools, systems, and practices) taking money from the unskilled. The same occurs in other walks of life, e.g. celebrity performers (movie stars, athletes, tycoons etc) make vast amounts of wealth relative to those who narrowly ‘miss the boat’. Moreover (in accounting for the high percentage of failed traders) transaction costs actually make trading a negative sum game, and (as can be shown with casino games) a small ‘edge’ will have a significant effect across a large enough number of trials, i.e. the millions of trades that take place every day. This has the effect of further increasing the gulf between the pro and the beginner. 8. With regard to those selling books, systems, tuition etc, in the case of those who do have a track record to vindicate their doing so, the answer to the question ‘why don’t you simply continue to grow your account exponentially with further trading?’ is that trading is a lonely, stressful and emotionally unrewarding pastime, and that they want to share their formula with others, whether for altruistic reasons, or to gain the adulation and commendation that they feel they deserve. But their pride tells them that they are in the 5% elite, and therefore that knowledge and expertise should be dispensed at a price. Meanwhile, the con-artists who sell dubious, untested material, without a comprehensive track record of their own, are giving TA a bad name, by exploiting the gullible majority. 9. With regard to fundamental analysis, technical analysts believe that price (and volume) are the only true and objective measure of what is actually happening, that the ultimate measure of crowd behavior is the MARKET ITSELF. They want to gauge the effect that news, fundamentals, etc has on the prices, before they make a decision. In their opinion, this makes fundamentals largely irrelevant. 10. Those who debunk TA are either too narrow-minded, lazy or unintelligent to perform the necessary research, and scoff for reasons of jealousy and pride. They would rather criticize from a safe distance, than participate in the heat of the boiler room, to discover whether or not they can make TA work for themselves. 11. Finally, and most importantly, TA adherents maintain that however unpredictable human emotions might be in responding to ‘externals’ such as news, trends and patterns will almost always exist, because the underlying forces driving price action are the irresistable sentiments of greed and fear. When participants see a price rise, they want to profit from the rise, so they buy the stock, and their keenness drives the price higher. When the price reaches a level where the majority become afraid of loss, they start to sell their positions, and crowd keenness to sell drives prices lower. This recurring cycle will always ensure the presence of trends, however erratic or short-lived they might be. [NOTE – With the sources and tools that are now available, there is perhaps a good argument for combining conventional TA techniques (trendlines, chart formations, candle patterns, indicators that are all based around OHLCV) with INDEPENDENT confirmation (e.g. space-time analysis, advanced pattern matching, genetic algorithms, astrophysics, astrological forecasting). However, die-hard TA proponents may register skepticism at approaches that have a ‘supernatural’, as opposed to ‘technical’, basis.]
Because it can be shown that (a) there are apparent cyclical patterns, but also apparent randomness, in price movements, and (b) most of the criteria above (especially the psychological factors) can not be objectively measured, it is close to impossible to determine which of the two arguments is correct. In my view, there is plausible evidence supporting both sides, although (if we assume that there are those who are making consistent profit, year-in-and-out), I suspect that they have placed enough trades to have attained ‘statistical significance’, i.e. that there is more than just ‘luck’ involved. Whatever the case, if we assume that is only the elite few that do apparently succeed, then (given the size of even the local marketplace) it is reasonable to infer that their gains must necessarily be astronomical. It is up to each individual to weigh up risk and return on the grandest possible scale, and ultimately decide whether he or she has the intelligence, the patience, the discipline, the nerve, and the resilience, to succeed at what must be one of the most demanding pastimes ever devised. That is the spirit shared by every entrepreneur, whatever his or her chosen walk of life.
See the following:
All of TA hangs on the underlying premise that elements of the past must, on balance, be more likely to recur in the future, than not. Otherwise, any kind of analysis would be futile, and it would be impossible for anybody to systematically profit from the markets. Put another way, if price movements were completely random, there would be exactly a 50-50 chance of guessing imminent direction, which means that, given enough time, transaction costs would eventually bleed even the luckiest guesser to death. Any kind of risk management would be futile; it would be like trying to beat the casino long term at Roulette. Trend following systems make the assumption that a trending component exists somewhere within the randomness, and that it can be identified and exploited, frequently enough, to overcome costs. This gives the expert trader his potential edge. Apart from commissions and slippage, the trend-follower forfeits profit because he is unable to buy at the absolute low, nor sell at the absolute peak. The edge must be large enough to, on balance, overcome all of this. And the pro who trades for a living faces another barrier: profit must occur consistently enough to meet practical and lifestyle considerations. Granted, any underlying math is imprecise, which can arguably put a "discretionary" system on an similar footing with a completely "mechanical" one. But that does not mean that, however deeply buried, it does not in some way exist. Consequently, historical testing over a large, statistically significant database of prices, can be used to approximately evaluate the effectiveness of a system of entries and exits. Note the following quote from market wizard Gil Blake: http://www.incrediblecharts.com/forums/messages/11/295533.html#POST43974 The fact that "history only repeats itself when it does" is what makes risk management imperative.
[This message has been edited by david_louisson (edited 12-17-2005).] IP: Logged |
|
mayur unregistered |
posted 01-13-2005 05:39 AM
Hi David, I was looking for some technical analysis software. Look what i found? I found best chart Plus your detailed article.So I have to look no further. Nice work. Keep posting. Mayur IP: Logged |
|
david_louisson Member Posts: 303 |
posted 01-29-2005 12:56 PM
ESSAY # 14.1 ===== INDEX OF ALTERNATIVE DATA SUPPLIERS ===== NOTE: I offer no recommendations as to data or product quality or reliability. I leave it to readers to conduct their own investigations. See Essay # 15.2 re using MSN Money Central as an alternative data source for historical charting in Best-Charts. http://moneycentral.msn.com/detail/stock_quote
American Stock Exchange
These have, for reasons unknown to me, apparently been deleted by website administrators. Some attempted humor, then, to replace the deleted posts:
12) Your 6-year-old pleads with you to take him to MACD's, and you ask him what the parameters are. 11) A social worker is telling you about a patient who has RSI, and you interrupt to ask her if she's read Wilder's book. (Then there's this patient with a history of volatility....) 10) An MA is no longer a university degree. 9) Trapped in traffic at a roundabout, you find yourself waiting for a "breakout". 8) You're constantly losing at tic-tac-toe because you keep visualizing it as a P&F chart. 7) A party addict is describing his LSD trips, and you ask whether his most recent high took out the previous one. 6) You describe an uneventful Friday at the office as an "inside day". 5) The best that lingerie advertisements can do is start you thinking about double tops. 4) While viewing the night sky with your hot date, you find yourself mentally constructing trendlines through the stars. 3) Your wife tells you she has PMT, but you can't remember what indicator that is. 2) You start thinking about your marriage in terms of risk-reward. 1) While engaged in, um, nocturnal recreation, you find yourself waiting for an entry signal.
[This message has been edited by david_louisson (edited 01-21-2006).] IP: Logged |
|
david_louisson Member Posts: 303 |
posted 02-06-2005 10:57 PM
Essay # 16.1 ===== BEST CHARTS MENU AND TOOLBAR OPTIONS ===== The following is a brief synopsis on each of Best-Charts’ (BC) menu options. NOTES: 1. Full help text is available from within BC (menu HELP > HELP CONTENTS > …).
FILE > CLOSE FILE > EXIT
MARKETS > U.S. MARKETS > (several options) MARKETS > STOCK FORUMS MARKETS > INVESTOR LINKS
PORTFOLIOS > LIST 1..20 PORTFOLIOS > ANALYZE 40 STOCKS (various options) Note that analyzing 40 stocks can take several minutes. When the process is complete, an HTML file (C:\M-c\TAsummary.html) showing the generated results is displayed. See Essay # 7.1 for information on how to interpret this output. PORTFOLIOS > RELOAD QUOTES PORTFOLIOS > INTRADAY MICROCHARTS PORTFOLIOS > TA CHARTS PORTFOLIOS > TECHNICAL ANALYSIS PORTFOLIOS > SAVE HIST OR INTRADAY QUOTES IN ASCII FILES ============== These parameters apply to all stocks. To set up different parameters for each stock, and for more detailed information on parameters in general, see Essay # 1.4 To select which one of the four parameter sets to use, use STOCKS > OPTIONS > Indicator Parameter Set. All subsequent TAs will use this parameter set. The parameter set used is shown in lavender typeface near the top right of the TA charts page (e.g. ‘Parameters: Set 1’). STOCKS > CHANGE WEIGHTS FOR BBI This function is useful in several situations, e.g. (1) if you want to base a heavily TA around a single indicator, without ignoring the others completely; (2) if you want to exclude an indicator from consideration altogether, because it is giving poor, erratic or unrealistic signals. STOCKS > SIGNAL FORECAST AND ESTIMATION STOCKS > COMPANY NEWS STOCKS > INTRADAY TA CHARTS STOCKS > READ END-OF-DAY (EOD) QUOTES FROM ASCII FILE (1) In STOCKS > OPTIONS > End of Day Quote File Path, enter the path here. (2) In STOCKS > OPTIONS> End of Day Quote File Path, select either ‘Specified file path 1’ or ‘Specified file path 2’. Then run STOCKS > SET END OF DAY QUOTE FILE PATH > SET FILE PATH 1 / 2, and type the path and file name. For example, if you enter STOCKS > SET END-OF-DAY QUOTE FILE PATH > SET FILE PATH 1 / 2 STOCKS > READ INTRADAY QUOTES FROM ASCII FILE (1) In STOCKS > OPTIONS > Intraday Quote File Path, enter the path here. (2) In STOCKS > OPTIONS> Intraday Quote File Path, select ‘Specified file path’. Then run STOCKS > SET INTRADAY QUOTE FILE PATH, and type the path and file name. For example, if you enter STOCKS > SET INTRADAY QUOTE FILE PATH STOCKS > READ HISTORICAL QUOTES FROM ASCII FILE (1) In STOCKS > OPTIONS > Historical Quote File Path, enter the path here. (2) In STOCKS > OPTIONS> Historical Quote File Path, select ‘Specified file path’. Then run STOCKS > SET HISTORICAL QUOTE FILE PATH, and type the path and file name. For example, if you enter STOCKS > SET HISTORICAL QUOTE FILE PATH READ HISTORICAL QUOTES FROM WEBSITE STOCKS > INPUT URL FOR READING QUOTES FROM WEBSITES STOCKS > SWITCH BETWEEN HISTORICAL QUOTES AND CHARTS STOCKS > DISPLAY CANDLESTICK CHART SIGNAL LIST STOCKS > BACK-TESTING (BT) Inputs: Buy/Sell Price – choose one of the following. This determines the price for each day that BT/AIO will use for buying and selling, from which profitability is calculated. Long / Short – choose one of the following. This determines whether BT/AIO will include long positions (trading for rising prices), short positions (trading for falling prices), or both. Performance Index – choose one of the following. This determines the basis on which BT/AIO will choose the most profitable calibration (i.e. parameters) for each indicator. Indicators – choose one of the following. This determines the indicator for which a complete history of trades will be shown in the back-test output. STOCKS > ALL INDICATOR OPTIMIZATION (AIO) > PARAMETER OPTIMIZATION Inputs: Check the indicators for which you want AIO performed. AIO is performed using the same inputs entered for BT (see previous section). STOCKS > ALL INDICATOR OPTIMIZATION (AIO) > DISPLAY OPTIMAL PARAMETERS STOCKS > ALL INDICATOR OPTIMIZATION (AIO) > SAVE OPTIMAL PARAMETERS STOCKS > ALL INDICATOR OPTIMIZATION (AIO) > USE OPTIMAL PARAMETERS STOCKS > ALL INDICATOR OPTIMIZATION (AIO) > DELETE OPTIMAL PARAMETERS STOCKS > MULTI-AIO STOCKS > SAVE ANALYSIS RESULTS IN RESULTS.TXT STOCKS > U.S. STOCKS > (various options) STOCKS > OPTIONS Indicator Parameter set: allows you to select Parameter Set 1,2, 3 or 4, when running unoptimized charts. See relevant section ‘STOCKS > CHANGE PARAMETERS OF INDICATORS > PARAMETER SET 1,2,3,4’ above, and also Essay # 1.4. End of Day Quote File Path: sets the path in which BC will search for EOD quote files. See section ‘STOCKS > READ END-OF-DAY (EOD) QUOTES FROM ASCII FILE’ above. Intraday Quote File Path: sets the path in which BC will search for Intraday quote files. See section ‘STOCKS > READ INTRADAY QUOTES FROM ASCII FILE’ above. Historical Quote File Path: sets the path in which BC will search for Historical quote files. See section ‘STOCKS > READ HISTORICAL QUOTES FROM ASCII FILE’ above. Analyze 200 or 400 Data: allows TA of either 200 or 400 data points (i.e. quotes or price bars). This sets the size of the base data window that is downloaded from Yahoo, and across which BT/AIO is constrained to operate. If you are back-testing across a large number (e.g. > 120) of quotes, then you will need to ensure that the data window is large enough (i.e. 400) to allow BT/AIO to generate meaningful results. Historical and/or Today's Quotes: Set to either ‘Hist. and Delayed Quotes’ or ‘Only Hist. Quotes’. Online or Offline: Set to ‘Online’, unless you are planning to do all analysis with local files (i.e. off-line from the Internet)
WINDOW > CASCADE WINDOW > TILE WINDOW > ARRANGE ICONS WINDOW > (window name)
HELP > ABOUT BEST-CHARTS HELP > BEST-CHARTS.COM / STOCK-ANAL.COM WEB SITE
1. (blank page icon), shortcut to FILE > NEW ===== END OF POST ===== [This message has been edited by david_louisson (edited 02-06-2005).] IP: Logged |
|
Admin Administrator Posts: 1091 |
posted 03-21-2005 11:04 AM
reopened IP: Logged |
|
david_louisson Member Posts: 303 |
posted 12-31-2005 11:09 PM
Happy New Year to all! I wish everybody prosperous trading in 2006. Essay # 5.3 has been added to this thread. David IP: Logged |
|
frankenstein Member Posts: 188 |
posted 01-01-2006 10:01 AM
Thanks again, David & Happy New Year to youn & yours.... frankenstein IP: Logged |
|
doug parkway Member Posts: 19 |
posted 01-26-2006 12:09 PM
Great to see a fellow kiwi on here mate...even if he is a Chiefs fan! I've got some information I'd really like to send you- drop me an email: glen@alphatrade.com All the very best IP: Logged |
|
mortsta Member Posts: 2 |
posted 02-09-2006 07:03 AM
David (and others) thanks so much for all the effort you have gone to explain the intracasies of BC...its greatly appreciated-if however my head is gonna explode! Each day i return and plod thru tom figure out what is meant...im getting there! BTW you kiwis...dont get too cocky re the super 14...your gonna get smashed by either the reds or the waratahs this year so look out...(hahaha) i cant decide who to support becuase i live in byron bay (NSW) and am 1/2 an hour from qld-HA!! anyway, if any of you kiwis are close by give me a shout-even if we beat you in the rugby! IP: Logged |
|
gerrydee Member Posts: 20 |
posted 04-27-2006 09:33 PM
David,is there a calculator that you know of that would facilitate the calculation of the R-value you speak of in essay 8.1? It really is tedious doing it by hand... IP: Logged |
|
david_louisson Member Posts: 303 |
posted 04-29-2006 06:18 AM
Hi Gerry I have tried doing a Google and found nothing. However, the formulae could be set up easily enough in Microsoft Excel. Does this help? David IP: Logged |
|
penniless Member Posts: 92 |
posted 05-25-2006 04:07 AM
David How difficult is it to link the subject index to the article so we can jump around. Thanks IP: Logged |
|
david_louisson Member Posts: 303 |
posted 05-29-2006 03:29 PM
Penniless I am not the webmaster. I can't rearrange the articles or provide additional references or hyperlinks beyond what I've already done. Perhaps you could copy and paste the whole thread into a suitable Word Processor and create your own bookmarks? IP: Logged |
|
penniless Member Posts: 92 |
posted 05-30-2006 03:08 PM
DAvid or Administrator Your following statement; How do you get and enter quotes from other sources. IP: Logged |
|
Admin Administrator Posts: 1091 |
posted 05-31-2006 10:11 AM
Choose the menu "St***s/Set URL to download quotes from websites" IP: Logged |
|
Arkady Member Posts: 8 |
posted 07-28-2006 05:27 PM
Hi, Look at the www.stock-forecasting.com This excellent site has 10 days and 6 month ahead prediction and email stock ALERT service! Actually I posted prediction for AMAT, Forecasted Data. AMAT www.stock-forecasting.com Date Open Close Low High Average Vector** Strategy*** Good luck!
IP: Logged |
|
simonDarf Member Posts: 1 |
posted 10-30-2006 04:47 PM
Theres lots of predictive sotware too, like matlab vantagepoint neuroXL many more. IP: Logged |
|
intech Member Posts: 47 |
posted 11-19-2006 06:17 AM
this link is not available. http://www.tradingblox.neturtles/turtlerules.pdf Book (PDF - needs Acrobat reader) outlining the original Turtle Trading Rules can be found here: IP: Logged |
|
david_louisson Member Posts: 303 |
posted 11-20-2006 09:19 PM
Intech Thanks for alerting me to this. I've fixed the link in Essay 6.2 to now read: I don't have the time to trawl the Internet regularly just to check whether all of the hundreds of links above are still available. If webmasters choose remove their links, there's not much any of us can do about it. What I suggest is that if you find some material that is particularly helpful, that you save it on your PC's hard drive. David [This message has been edited by david_louisson (edited 11-20-2006).] IP: Logged |
|
david_louisson Member Posts: 303 |
posted 12-12-2006 05:54 PM
New material has been added - for details, see the "What's New" section at the top of the thread. Hence I'm replying to my own post, to bump this thread back to the top of the forum. Happy reading IP: Logged |
This topic is 2 pages long: 1 2 All times are ET (US) | next newest topic | next oldest topic |
  |
|
Contact Us | Home Page
Powered by Infopop www.infopop.com © 2000
Ultimate Bulletin Board 5.46