Candlestick patterns can signal a trend continuation or reversal. Many traders and investors know how to benefit from candlesticks analysis. As a rule, a simple chart analysis is used that requires analysts to remember and correctly interpret, at least, typical candlestick patterns. However, the number of possible different candlestick shapes and their combinations in a row can be huge. Besides, some patterns interpretations might be considered as contradictory or doubtful.
Fortunately, statistical methods combined with computer power can be a good solution to make the candlestick patterns recognition works less time-consuming and more effective. These days, Neural Network (NN) can help to automate a candlestick patterns recognition task. NN should be properly trained in order to be able to predict the next candlestick parameters with the highest probability. One of the obvious problems of implementing a candlestick pattern NN predicting system is a formalization of inputs.
Numerous tests show many possibilities of improving NN candlestick patterns recognition systems. For example, output result can be composed from selected optimized calculations based on different historical periods. Also there are many different ways to formalize the shapes and relative positions of candlesticks. As well as, as back- and forward-testing show, it is reasonable to re-train NN for each particular type of shares and latest historical periods to make the forecast more accurate.
Optimal Solution. There is an automated tool FTA-2 (free use of fully-functional version for one month). It has module which enables using Neural Network to recognize typical candlestick patterns and predict future prices (open, high, low, close). This module predicts only one next candlestick but the prediction can be successfully used for different widths of candlestick, i.e., the number of trading days in one candlestick. The module has been enhanced to calculate result that is composed from different historical periods that allows making the forecast more accurate. Also it can perform comparative forecast analysis for many symbols.
- Candlestick basics – major signals
- Neural Network basics – introduction
- The computer program which enables using Neural Network to recognize typical candles patterns and predict future prices – software FTA-2
Categories: Research, Software automated, candles, candlestick, neural network, NN, pattern recognition, predict, reversal, statistical methods, stock market, trend
Parabolic SAR (SAR stands for Stop-And-Reverse) is a trend-following indicator that has been used by many traders for decades. Its major application is in trading systems to define a trailing stop, i.e., to protect profit when a price trend changes. The term “parabolic” appeared to characterize the indicator parabola shape that is due to using an accelerating factor in the formula. SAR is especially effective in a trending market. To make it more effective in a sideways market, it is often used in conjunction with other indicators.
SAR indicator gives a strong signal when a price trend is about to reverse, therefore, this indicator can be used for prediction. To compare the predictive ability of SAR with other indicators, it has been implemented into the technical analysis module of Fundamental-Technical Analyzer FTA-2. SAR calculations have been used to collect statistics based on the forecast simulations for major indexes and ETFs during August-October 2011 period. As a result, SAR’s position was mostly in “top ten” indicators list.
The research and presented chart are made by Fundamental-Technical Analyzer FTA-2, one of the software modules that enables composing Neural Network forecasts of many indicators with weights accordingly to each indicator’s predictive ability.
Omitting logical rules for accelerating factor and reversal conditions, a recurring core formula for SAR is the following:
SAR (current point) = AF * [EP – SAR (previous point)] + SAR (previous point)
AF – Acceleration Factor (normally starts from 0.02 and increases by 0.02 if each next point reaches a new extreme, saturates until 0.2);
EP – Extreme Point (lowest low or highest high).
To summarize, Parabolic SAR can be enriched by combining it with Neural Network and successfully used for predicting stock market prices. Combing it with Neural Network allows extracting more statistically stable patterns and, therefore, providing a better accuracy in the forecast. As simulations showed, improved results can be achieved if SAR is transformed into more sensitive indicator by subtracting it from close price (it indicates the degree of SAR and price convergence).
Some indicators can provide a better prediction than others so that it seems logically to use the best selected ones to build a composite forecast. On the other hand, even the best indicators can fail. The questions is how to get a consistent good accuracy in predicting – by using only a few best indicators or many good ones. The answer is not obvious and, therefore, a factual comparative analysis would be needed to shed some light on this issue. This short report is based on limited statistical researches; it is an attempt to reach a certain conclusion.
About Expert Method. Apparently, the more good forecasts are taken into consideration, the more precise can be an approximation to actual value. There is Expert Method. This method can be explained by following. As example, an experimentalist shows a pen and asks a group of about 40 people to write down their estimate of the length. Then he collects notes and calculates the average number – normally it is almost 100% accurate. Why it works? Because everyone makes errors in different directions so that averaging gives a precise result.
The Details of Experiment. To find an optimal number of top performing indicators, two tests have been done – using artificial data and real market data. Artificial data allow performing forward testing with more consistent statistics. Although back-testing has been done on out-of-sample sets, it did not have the same forward-testing success every time. Forward testing showed that in average few indicators might produce less accurate prediction than many.
The researches and presented chart are made by Technical Analyzer TA-1 (the software is able to compose Neural Network forecasts of many indicators with weights accordingly to each indicator’s predictive ability).
Conclusion. The main conclusion is that relying on a couple of best indicators might yields less consistent success over a long run than using many best and good ones. However, too many is another extreme and not good. The second conclusion is that the list of best indicators is not static – it evolves depending on many factors, including market conditions and, probably, on the number of traders employing particular indicators to make their buy-sell decisions. Thirdly, if the best current top of indicators is known, here is a magic number – it is around 30. And finally, better results are possible if indicators are combined accordingly to their latest back-testing ranking.
Technical analysts use many different indicators. Not all indicators are equally good. Some of them have better predictive abilities than others at given conditions. Statistical results of the research for stocks and indexes during the last several months have showed the way to improve some regular indicators. In short, if an indicator is trend-differentially coupled with price – it demonstrates better predictive abilities than a pure indicator.
Among studied indicators are Relative Strength Index (RSI) and Moving Average Convergence/Divergence (MACD) indicators. They have been transformed to a slope of line and differentially-coupled with a price line slope. Indicators and price transformation to line slopes has been performed using Least Squares Linear Regression within a sliding 10-day period (moving window). A comparative analysis of indicators’ predictive abilities showed that these two coupled ones are better than around 90% of all other (57) the most popular indicators.
The chart below shows an example of such forecast:
The research results presented here have been calculated using a feature of Technical Analyzer TA-1 (TA). Except analyzing chart with indicators and historical data, TA enables to perform a 10-day forecast using Artificial Neural Network. The calculations can be done on the basis of one selected technical indicator or all available. If all indicators used, TA decides how much weight should be assigned for each indicator’s forecast in a composite result by using back-testing for particular market conditions and a specific stock. Each weight is proportional to a predictive ability of a corresponding technical indicator.
The new group Trading Decision Support Systems is intended to be a resource for individual/institutional traders/investors and software developers in stock market area to share ideas, initiate and participate discussions, benefit from the collective intelligence, and to expand network. It will be primarily focused on such topics as:
- Trading EOD and intraday different asset classes: trading tips, strategies, why, how, and results.
- Trading systems: algorithms, methods, technologies, human factor, and statistics.
- Software tools to support traders decisions: forecast methods, simulations, back-testing, and optimization.
- Technical Analysis: indicators and chart patterns.
- Fundamental Analysis: financial ratios and predictive models.
- News: analysis and formalization by converting to measurable variables to automate systems with contributing news factor.
- Numerical methods, data processing, artificial intelligence, and modeling in stock market areas.
Many things remain unchangeable in a trading world – supply-demand price balance, greed-fear driven mistakes, as well as, ability to think, make right decisions, and find the best solutions. When once winning approaches, strategies, or methods failed, many traders are prone to analyze the reasons why it happened. Then they create new approaches and develop new successful systems. If systems are automated, it is easy and fast to test them, collect and analyze back-testing and live statistics, and then make necessary improvements. That is why it is important to implement the best ideas in software applications that can be also used by others.
The computational technologies are changing. Systems empowered by Artificial Intelligence have self-learning abilities that enable them to adapt to market changes. One of the purposes of this group is to bring together the developers of decision support software and traders-users for mutual benefits: the developers get more ideas about their products’ improvements and make a better progress in developing software for traders, the users arise issues relating to their needs and wants. Hopefully everyone will find something useful participating in this group.
You are welcome to join this newly created networking group. Be the first to start a relevant discussion, promote your product or service. Please join Trading Decision Support Systems group on LinkedIn!
Stock market investors and traders are making their decisions trying to analyze all relevant factors. Indeed, a good decision should be based on a sufficient number of criteria taking into consideration many company’s fundamentals, its stock characteristics, and external factors. During this process the market participants dealing with selecting sufficient volume of data and converting quantitative and qualitative parameters into valuable information. How to deal with a huge amount of data? How to choose the right data to analyze? What is the best way to process all info inputs?
One of the parts in decision-making chain that is often overlooked is a proper interpretation of data. There are different risks of converting data into valuable information. One of them occurs when some of the factors are exceedingly overvalued and others are neglectfully undervalued and, therefore, this might significantly diminish the quality of output signals for making decision.
One of the solutions is to use a computer program Investment Analyzer InvAn-3/4. It processes automatically all input data converting them into one number, i.e., composite rating. For better analysis, important company-stock characteristics are divided into homogeneous groups as components for further calculation of the composite rating. These components are: fundamental, technical, and timing ratings (FTTR). Finally, the program combines FTTR that allows measuring the quality of a company-stock by a single number.
The composite rating is a result of harmonic averaging of FTTR that initially transformed by Quality Auto-Function (HAQF method) with different weights. Applying HAQF to FTTR allows modeling stock and its company quality very realistically. It enables to reflect accurately the quality of company-stock as a system. In other words, this composition algorithm of the components ensures the following three principles:
- System reliability depends on reliability of its weakest part(s)
- Too good some system’s component(s) cannot mask the rest of components
- The composite result is formed by all components accordingly to each component’s importance
The synthesis of ratios and parameters is performed on the basis of HAQF method that allows modeling company-stock state and dynamics very close to reality. Quality Function (QF) transforms these ratios and parameters (factors) with different critical and sufficient values. The main purpose of Quality Function is to model system quality by normalizing with two characteristic parameters – sufficient and critical. When a certain parameter has its value less than critical, the function approaches to zero. If some parameter exceeds the sufficient value, the function ceases increasing. The following example of non-symmetrical Quality Function (critical value is 30, sufficient value is 250) is to help to understand QF:
The purpose of this research was to compare two statistical methods: one that based on Cycle Analysis, another – on a simple Neural Network. Price and volume data were used to train this particular Neural Network. These statistical forecasts were built using historical data of S&P-500 index for six months (from June 2009 to January 2010).
The charts below shows how actual 5-day performance (yellow line) differ from predicted performances by these two methods. The top half is the comparison of Neural Network prediction, bottom half – Cycle Analysis. Green bars mean buy signals, red – sell*.
Three major conclusions for this particular historical period:
- Cycle Analysis prediction gives signals too early, Neural Network prediction – too late.
- In average, the prediction by Cycle Analysis showed slightly better accuracy than the one by Neural Network.
- It seems logical to combine these two methods to improve the accuracy.
* ) The calculations have been performed by an integrated experimental system that combined two applications: Cycle Analysis predictor SMAP-3 and Neural Network predictor NNSTP-2
Chart represents S&P-500 forecast for August 24 – 28, 2009. The calculation has been performed using Neural Network Stock Trend Predictor NNSTP-2 (input data – price and volume).
Case Study: Making Investment Decision
Making an investment-related decision involve gathering data, analysis, and prediction. As a rule, at any moment, two groups of factors exert influence on the decisions – positive and negative. To minimize the investment risk, all factors should be properly evaluated. This article shows an example of a real time experiment of making investment decisions. The experiment started in February 2009. The initial amount of fund for investing was around USD 2000. Below are some inputs for making the first buying decision at that time:
- The US and global economies were in a bad shape. Although the stock market showed some weak signs of recovering, most investors considered the market as bearish.
- In addition, there was some risk for the market to decline in the short term mostly due to expected disappointing financial reports for the fourth quarter of 2008.
- On the other hand, normally, month February is a annual cyclical minimum for energy sector stocks. It was confirmed by the chart of the indexes that represents the energy sector (XLE and XEG.TO). The result was calculated by SMAP-2 using the recent 12-year period.
- Statistical research may not be accurate if something new suddenly appears.
- Another positive factor was the fact that the equity markets are leading indicators of economy. As a rule, the stock market starts recovering in around six months in advance before the economy.
- Calculated composite rating (by InvAn-4) and comparative analysis of sectors showed a favorable position for energy sector stocks. OII was one of the stocks from energy sector with high composite rating. OII has all good three components of composite rating – fundamental, technical, and timing ratings.
- On February 6th, the five-day forecast by NNSTP-1 showed some uptrend so that the entry point (buying) was chosen correctly. The stock OII was purchased.
After purchase the overall stock market moved slightly up and then went down more erasing previous gain. The lesson – a market correction (decline) can be used as an opportunity to maximize return. However, eventually the stock moved up much more with solid ROI – more than 25% in April.