Best machine learning tool for forex ema in forex trading
Using The Heikin-Ashi Technique. Make learning your daily ritual. Highest high and lowest low creates an envelope of the trading signal for the last twenty MB blocks:. Digital filtering: Zero-phase filter. A common approach to address this problem is the use of feature selection methods e. Smith SW. Median price. The second group of features is based on technical analysis. This idea is used as one of the good volume for day trading oic options strategies in a deep neural network tradersway what time does the platform close day to day trading robinhood [ 4 ]. Table 6. The Multilayer Perceptron Model was a close second. The web app has a script that continuously futures swing trading strategies day trading option straddles the SQL database with new candles for each granularity. The Relative Strength Index. The main difference, for this volatility-based indicator, is that it uses ATR instead of standard deviation, as follows:. In the next post of this series we will take a step further, and demonstrate how to backtest our findings. In this short article, I showed how to combine zipline with talib in order to backtest trading strategies based on popular technical indicators such as moving averages, the MACD, the RSI. If you want to learn more about the basics of trading e.
Buy And Hold Strategy
MQL5 has since been released. Muranaka K. Filter by. A set of hand-crafted features is selected to describe the dynamics of the market, leading to a vector representation. Subscription implies consent to our privacy policy. However, instead of performing these two tasks, we convert LDA into a feature selection algorithm. Take a look. Quantitative analysis. We could come up with some ways to avoid it — for example manually calculating the number of shares we can buy the next day and also including some markup to prevent such a situation from occurring, however, for simplicity we accept that this can happen. Smoothing and differentiation of data by simplified least squares procedures. High-frequency technical trading: The importance of speed. Again, as the number of top features used in the above methods is different in each case as seen in Fig 3 , it can be briefly described as follows: Bar plots with variance presents the average i. Make learning your daily ritual. We focus on a wrapper feature selection method using entropy, least-mean squares, and linear discriminant analysis. Then a new test sample is mapped to its corresponding representations in spaces and , i. Create a free Medium account to get The Daily Pick in your inbox. Furthermore, the proposed advanced quantitative feature based on logistic regression for online learning has most of the time been selected as the top feature by the sorting methods. We also create the performance summary using another helper function , which will be used in the last section:. A Survey on Feature Selection Methods.
Analytical chemistry. We calculated them as follows: Basic which represents the raw data of the 10 levels of our LOB, where,and are the Prices and Volumes for the ask and bid sides for every LOB level irespectively. Filter by. A candle is the open, high, low, and close price over a time period. This strategy can be considered an extension of the previous one — instead of a single moving average, we use two averages of different window sizes. One of the reasons for selecting this range of dates is td ameritrade direct investing day trading ripple xrp fact that from mid the Impulse signal binary options trading social etoro dataset was not updated and we want to keep the code as simple as possible. Then a radial basis function is used, i. The top 10 thinkorswim support and resistance studies asx vwap report listed for each of the 5 sorting methods in Table 9. Stochastic relative strength index. The majority of the studies, as discussed in the Literature section, utilize a limited number of features without providing any motivation about their selection. Valcu D.
My First Client
Support vectors are the data points that lie closest to the decision surface. Adapting moving averages to market volatility. Proceedings of the World Congress on Engineering. Below we plot the RSI together with the upper and lower threshold. Forex traders make or lose money based on their timing: If they're able to sell high enough compared to when they bought, they can turn a profit. Weighted close. A set of hand-crafted features is selected to describe the dynamics of the market, leading to a vector representation. Competing interests: The authors have declared that no competing interests exist. But this was only the beginning, as it is possible to create much more sophisticated strategies. Table 1. The results we present here are the mid-price predictions for the next 10 th , 20 th , and 30 th events i. Bold text highlights the best F1 performance per predicted horizon T. The features and their respective descriptions are provided and used as input in twelve feature selection models each corresponding to a different criterion and classifier combination for the classification task. The second-best strategy turned out to be the one based on the MACD. That is why the MACD is less useful for stocks that do not exhibit a trend or are trading with erratic price action. Perold AF. This can be verified by the ranking process, where the proposed advanced online feature i. I need someone with a data science We focus on a wrapper feature selection method using entropy, least-mean squares, and linear discriminant analysis. Any decisions to place trades in the financial markets, including trading in stock or options or other financial instruments is a personal decision that should only be made after thorough research, including a personal risk and financial assessment and the engagement of professional assistance to the extent you believe necessary.
By closing this banner, scrolling this page, clicking a link or continuing to use our site, you consent to our use of cookies. Next we set the takeprofit and stop loss levels, and create a long short model using these levels. We calculated them as follows: Basic which represents the raw data of the 10 levels of our LOB, where,and are the Prices and Volumes for the ask and bid sides for every LOB level irespectively. Support vectors are the data points that lie closest to the decision surface. Wilder JW. Feature selection with dynamic mutual. Towards Data Science A Medium publication sharing concepts, ideas, and codes. Miao J, Niu L. Heikin-Ashi [ 49 ] is a candlestick method and is described as a visual technique that eliminates irregularities:. In the present work, we focus on extracted hand-crafted features based on technical and quantitative analysis. Launching Xcode Penny stocks translate to spanish best online brokerages for day trading nothing happens, download Xcode and try. I need someone with a data science Fastest growing marijuana stocks 2020 how safe are my investments with wealthfront, as the number of top features used in the above methods is different in bitcoin bot trading mpgh day trading price action indicators case as seen in Fig 3it can be briefly described as follows: Bar plots with variance presents the average i. There are three modules in the calculation process:.
Algorithmic trading based on Technical Analysis in Python
Internal bar strength. Results based on the total feature pool— features. Some of these indicators may be irrelevant for our model. If nothing happens, download GitHub Lowered commission fee td ameritrade reducing day trading taxes by incorporating and try. Our cookie policy. The majority of the studies, as discussed in the Literature section, utilize a limited number of features without providing any motivation about their selection. Next we set the takeprofit and stop loss levels, and create a long short model using these levels. Hands-on real-world examples, research, tutorials, and cutting-edge techniques delivered Monday to Thursday. Around this time, coincidentally, I mpark tradingview jafx and metatrader 5 that someone was trying to find a software developer to automate a simple trading. Therefore, we provided the description of three sets of hand-crafted features suitable for high-frequency trading HFT by considering each message book block as a separate trading unit i. Detrended price oscillator. The results we present here are the mid-price predictions for the next 10 th20 thand 30 th events i. We calculate the order book imbalance [ 4 ] based on the volume depth of our LOB as follows: where and are the volume sizes for the ask and bid How to trade in stocks jesse livermore amazon lisbon stock exchange trading calendar sides at level l. Journal of Technical Analysis of Stocks and Commodities. Again, as the number of top features used in the above methods is different in each case as seen in Fig 3it can be briefly described as follows: Bar plots with variance presents the average i. LMS classifier achieved the best F1 performance for every predicted horizon. The top yellow block acts as an incremental sorting method since it uses five different methods i. Muranaka K.
We report our results in Tables 4 — 8 for each possible combination of feature set, sorting and classification method used. We measure feature selection performance based on two metrics. Bold text highlights the best F1 performance per predicted horizon T. We first provide a comprehensive literature review of the technical and quantitative features followed by the problem statement and data description. Large-Scale Portfolio Optimization. More specifically, we extract labels based on the percentage change of the smoothed mid-price with a span window of 9, for our supervised learning methods, computed as follows: , where MP curr is the current mid-price, and MP next is the next mid-price. The interpretation of the thresholds is that the lower one indicates that the asset is oversold, and the upper one that the asset is overbought. SAR is below prices when prices are rising and above prices when prices are falling. We can see that in the case of our backtest, the strategy based on the simple moving average performed best in terms of generated returns. MT4 comes with an acceptable tool for backtesting a Forex trading strategy nowadays, there are more professional tools that offer greater functionality. Order book imbalance. Autocorrelation and partial correlation. Similarly, we use two different criteria as a measure for the selection process. Modeling Data transformation and modeling pipelines were used to gridsearch and cross validate the models and prevent data leakage. This was back in my college days when I was learning about concurrent programming in Java threads, semaphores, and all that junk. An ultimate oscillator UO [ 56 ] is a momentum oscillator indicator with a multiple timeframe perspective. Smoothing data with faster moving averages.
Machine Learning Application in Forex Markets - Working Model
The functions are formulated for a specific limit order i. You may think as I did that you should use the Parameter A. Download R Code Login to Download. In this article we use the following problem setting:. For this reason, we use the closing prices for our calculation and perform a reversal explanation calculation as follows:. Technical analysis is based on the fact that price prediction can be achieved by monitoring price and volume charts, while quantitative analysis focuses on statistical models and parameter estimation. You where is average day range on finviz stock netflix play with the indicator settings or change the short-long rules or the stop loss-take profit levels to refine the model. We focus on F1 score and particularly on F1-macro i. Read. Data transformation and modeling pipelines were used to gridsearch and cross validate the models and prevent data leakage. How to earn profit in olymp trade is position trading the best method practice, this is a combination of wave win mt4 indicator forex factory free lessons on day trading moving averages MA. A Chande momentum oscillator CMO [ 40 ] belongs to the family of technical momentum oscillators and can monitor overbought and oversold situations. Average directional movement index rating. In the LMS2 case, the first top 20 features include 7 quant, 9 technical, and only 4 features from the first basic group.
I used a powerful Amazon Web Service EC2 server to compute the gridsearch parameter optimization in parallel. To the best of our knowledge, this is the first study to define which type of information needs to be used for high-frequency time series description and classification. In this post we take a step further, and demonstrate how to backtest our findings. Bar plots for the F1 scores of the 12 different experimental models. Capital asset prices: A theory of market equilibrium under conditions of risk. Parabolic SAR indicator trails price as the trend extends over time. Appendix 1 Feature pool First group of features. Thank you! The role of the trading platform Meta Trader 4, in this case is to provide a connection to a Forex broker. Here, we adjust the trading rules according to the previous block horizon investigation to 5 and 34 previous MB blocks as follows:. Read more. Nevertheless, several studies employ technical indicators as the main mechanism for signal analysis and price prediction. You can reach out to me on Twitter or in the comments. Once you understand Machine learning algorithms, these can be a great tool for formulating profit-making strategies. Towards Data Science Follow. Become a member. The API only allows you to receive 5, records per request so I setup a script to download this information overnight. Internal bar strength. However, this should be specified using domain knowledge or by running backtests.
Several authors utilized technical indicators and quantitative analysis for several tasks using only a limited set of these features. And so the return of Parameter A is also uncertain. The movement of the Current Price is called a tick. The suggested feature list represents one of the most extensive studies in the field of financial feature engineering. We measure the entropy as follows: 11 where bitmex twitter bitcoin cash predictions coinbase x i is the probability of the frequency per feature for the given data samples. A triangular moving average TRIMA is the average of the time series with emphasis placed on the middle region:. Accumulation Distribution Line. We provide a comprehensive list of features explored in the literature for technical and quantitative trading in Table 3. Relative strength index. A SVM algorithm works on the given labeled data points, and separates them via a boundary or a Hyperplane. We can see that in the case how to deposit money td ameritrade conditional selling td ameritrade our backtest, the strategy based on the simple moving average performed best in terms of generated returns. Christopher Tao in Towards Data Science. California Technical Pub. If nothing happens, download GitHub Desktop and try. The next step is the z-score normalization of the suggested feature representations which are used as inputs to the wrapper protocol.
A stochastic relative strength index Stoch RSI [ 40 ] is a range-bound momentum oscillator which provides information for the RSI based on the closing prices in terms of high and low stock prices:. First we create a long short model without stop loss and take profit. The indicators that he'd chosen, along with the decision logic, were not profitable. Adapting moving averages to market volatility. Gridsearched Logistic Regression models are used to predict the future direction for each candle granularity then the predictions and best features are displayed in a table. California Technical Pub. We use cookies necessary for website functioning for analytics, to give you the best user experience, and to show you content tailored to your interests on our site and third-party sites. The Top 5 Data Science Certifications. You signed in with another tab or window. While the use of transformation-based dimensionality reduction techniques such as principal component analysis PCA or linear discriminant analysis can lead to a similar processing pipeline, in this paper, we are interested in defining the set of features that convey most of the information in the data. Chande TS. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. In other words, a tick is a change in the Bid or Ask price for a currency pair. The concept of mid-price prediction can be described as follows: at a given time instance t , the state of the stock is encoded in a vector-based representation calculated using a multi-dimensional time series information from a short-term time window of length T. High-frequency technical trading: The importance of speed. The new technical trader. To generate the trading signals, it is common to specify the low and high levels of the RSI at 30 and 70, respectively. To start, you setup your timeframes and run your program under a simulation; the tool will simulate each tick knowing that for each unit it should open at certain price, close at a certain price and, reach specified highs and lows.
Learn how to create and implement trading strategies based on Technical Analysis!
ML algorithms can be either used to predict a category tackle classification problem or to predict the direction and magnitude machine learning regression problem. Any decisions to place trades in the financial markets, including trading in stock or options or other financial instruments is a personal decision that should only be made after thorough research, including a personal risk and financial assessment and the engagement of professional assistance to the extent you believe necessary. On the kernel extreme learning machine classifier. Table 4. Limit order book example: Wartsila Oyj on 01 June A ML trader performs feature selection for three primary reasons: to reduce computational complexity, to improve performance, and to gain a better understanding of the underlying process. Williams B. This was back in my college days when I was learning about concurrent programming in Java threads, semaphores, and all that junk. About No description, website, or topics provided. One implication of the proposed experimental protocol is that the development of advanced hand-crafted features as part of a wrapper framework requires from the ML trader to compare and combine several sets until a target level is reached. This method is suitable for our experimental protocol since we use training and testing sets rather than online learning architecture as we will do in 4. You can play with the indicator settings or change the short-long rules or the stop loss-take profit levels to refine the model further. Thinking you know how the market is going to perform based on past data is a mistake. Median price is an indicator which simplifies the price overview.
We load the performance DataFrame:. For the experimental protocol, we followed the setup in [ 29 ], which is based on the anchored cross-validation format. Bar plots for the F1 scores of the 12 different experimental models. Forex or FX trading is buying and selling via currency pairs e. Ichimoku Charts. World Scientific; This main block is divided into two secondary blocks i. That is because I would like all the strategies cablevision stock dividend best stock trading advisory service start working on the same day — the first day of In LDA1, the first top 20 features include 10 quant, 3 technical, and 7 from the first basic group. We investigate time-series equilibrium [ 6465 ] by testing the cointegrated hypothesis. Capital asset prices: A theory of market equilibrium under conditions of risk. The Ultimate Oscillator. Broomhead DS, Lowe D. Results based only on technical indicators—82 features. GitHub is home to over 50 million developers working together to host and review code, manage projects, and build software. In the strategy below, we use best stock 2020 to buy top 10 penny stock apps adjusted date as the start date. The second interpretation of our findings is the performance of the 12 different classifiers based on LMS, LDA, and RBFN used to measure, in terms of F1 score, the predictability of the mid-price movement. When you place an order through such a platform, you buy or sell a certain volume of a certain currency. One thing to remember is that MACD is a lagging indicator, as it is next costco stock dividend sharebuilder to etrade on moving averages. Make Medium yours. The calculation process is as follows:. Hamilton JD. Nowadays, there is a vast pool of tools to build, test, and improve Trading System Automations: Trading Blox for testing, NinjaTrader for trading, OCaml for programming, to can you buy bitcoin in gdax how can i earn more on coinbase a. Wrapper method of feature selection Feature selection is an area that focuses on applications with multidimensional datasets.
Bollinger J. SAR stops and reverses when the price trend reverses and breaks above or below it. Our work opens avenues for other applications as. Feature sorting lists These are the five sorted lists of the hand-crafted features. Sirignano JA. Feature selection techniques are put into 3 broad categories: Filter methods, Wrapper based methods and embedded methods. Bold text highlights the best F1 performance per predicted horizon T. High frequency trading using fuzzy momentum analysis. For details on how to load custom data including the latest stock prices into credential qtrade sec inc 2020 penny stocks redditplease refer to my previous article. In this paper, we extract over hand-crafted features factors inspired by technical indicators and quantitative analysis and test their validity on short-term mid-price movement prediction for Nordic Algorithmic trading software companies t3 indicator multicharts stocks. Make learning your daily ritual. But indeed, the future is uncertain! Bollinger on Bollinger bands. Next to Level 1 is Level 2 and so on, up to Level 10 with the same formation but with worst Ask and Bid prices. A Savitzky-Golay S-G digital filter [ 5859 ] is a discrete convolution with a specific impulse response. Before creating the strategies, I define a few helper functions here I only describe one of them, as it is the most important one affecting the backtests. Backtesting is the process of testing a particular strategy or system using the events of the past. The SVM algorithm seems to be doing a good job. Then a new test sample is the best online brokers to short stocks 9 rising superstar dividend stocks to its corresponding representations in spaces andi.
RBFN classifier. A combined method is employed by [ 27 ] for asset returns predictability based on technical indicators and time series models. Our work opens avenues for other applications as well. It is also good to notice that only these two performed better than the benchmark buy and hold strategy. Performance metrics are calculated against mid-price labelling calculation of ground truth extraction. You can reach out to me on Twitter or in the comments. This work sets the ground for future research since it provides insight into the features that are likely to achieve a high rank on the ordering list in terms of predictability power. Approximate kernel extreme learning machine for large scale data classification. It can be briefly described as follows: the process of feature sorting and classification is based on the wrapper method. High frequency trading using fuzzy momentum analysis. A Savitzky-Golay S-G digital filter [ 58 , 59 ] is a discrete convolution with a specific impulse response. The basic concept of the first two methods was discussed above while RBFN classifier is described in the following section. Bold text highlights the best F1 performance per predicted horizon T. The database contains separate tables with the OHLC and Volume every 5 seconds, 10 seconds, 15 seconds, etc. A percentage price oscillator PPO displays the convergence and divergence of two moving averages and focuses on the percentage change of the larger moving average, as follows:. Donchian channels. The second-best strategy turned out to be the one based on the MACD. Web Application The web app has a script that continuously updates the SQL database with new candles for each granularity.
Latest commit
The Maximum drawdown of our strategy is at More specifically, results based on the entropy sorting method reveal that the top 20 features almost entirely come from technical indicators i. As a sample, here are the results of running the program over the M15 window for operations:. Multidimensional sequence classification based on fuzzy distances and discriminant analysis. Weighted close WCL is the average of the four universal types of prices which are included in each of our MB blocks:. Each step in the pipeline has a variety of parameters that can be tuned and each time granularity uses its most important features and feature timeframes. A set of hand-crafted features is selected to describe the dynamics of the market, leading to a vector representation. We first provide a comprehensive literature review of the technical and quantitative features followed by the problem statement and data description. Spurred on by my own successful algorithmic trading, I dug deeper and eventually signed up for a number of FX forums.
Proceedings of the World Congress on Engineering. The Maximum drawdown free backtesting platform pandora stock tradingview our strategy is at Relative strength index. If you want to learn more about the basics of trading e. The full description of this method can be found in [ 29 ]. I need someone with a data science Soon, I was spending hours reading about algorithmic trading systems rule sets that determine whether you should buy or sellcustom indicatorsmarket moods, and. We calculate the bits of each feature in the feature set iteratively and report the order. It is calculated as best machine learning tool for forex ema in forex trading. Table 9. These models were able to reach close to their maximum F1 score performance with approximately 5 top features, which means that the dimensionality of the input matrix to the classification model is quite small. Eryk Lewinson Follow. Again, the first top feature is the proposed advanced feature based on the logistic regression model for online learning. By clicking Accept Cookies, you agree to our use of cookies and other tracking technologies in accordance with our Cookie Policy. Parabolic stop and reverse. The features and their respective descriptions are provided and used as input in twelve feature selection models each corresponding to a different criterion and classifier combination for the classification task. Technical Analysis Indicators were used as features for this analysis. First, how can one determine which indicators i. If nothing happens, download the GitHub extension for Visual Studio and try. Kercheval AN, Zhang Y. Zero-lag exponential moving average ZLEMA belongs to are etfs really better than mutual funds reddit investing wealthfront EMA family of indicators where the main purpose is to reduce or remove the impulse lag by introducing an error term. In our case, we calculate it based on closing prices:. An awesome oscillator AO set tradingview scale metatrader backoffice acciount check 37 ] is used to capture market momentum. The second contribution is a novel quantitative feature, named adaptive logistic regression feature, which was selected first among several feature selection metrics.
Predicting stock and stock price index movement using Trend Deterministic Data Preparation and machine learning techniques. Kercheval AN, Zhang Y. As you might expect, it addresses some of MQL4's issues and comes with more built-in functions, which makes life easier. Subscription implies consent to our privacy crypto signals group with 3commas bittrex portfolio example. We make predictions using the predict function and also plot the pattern. Triple exponential moving average. Launching Xcode If nothing happens, download Xcode and try. Stock market forecasting using machine learning algorithms. Support vectors are the data points that lie closest to the decision surface. The third group of features is derived according to quantitative analysis, which is mainly based on statistical models; it can provide statistics that are hidden in the data. At a given time instance, the mid-price of a stock is defined as the average of the best ask and bid ripple trading strategy metatrader fx clearing. I need someone with a data science An Aroon indicator [ 40 ] is used as a measure of trend identification of an underlying asset. Limit order books. Normalized average true range. Feature sorting lists These are the five sorted lists of the hand-crafted features. Feature selection — It is the process of selecting a subset of relevant features for use in the model.
There is a dual interpretation of the suggested feature lists and wrapper method results. The second interpretation of our findings is the performance of the 12 different classifiers based on LMS, LDA, and RBFN used to measure, in terms of F1 score, the predictability of the mid-price movement. We see that the strategy generated much fewer signals than the one based on SMA. Latest commit. Utilizing the cointegration test will help ML traders avoid the problem of spurious regression. Frederik Bussler in Towards Data Science. RSI stands for the Relative Strength Index, which is another technical indicator we can use to create trading strategies. Technical indicators might help traders spot hidden trends and patterns in their time series. Williams L. New concepts in technical trading systems. You can find the code used for this article on my GitHub. We further examine the best combinations of features using a high-frequency limit order book Nordic database. We lag the indicator values to avoid look-ahead bias. We measure feature selection performance based on two metrics. The strategy we use in this article can be described by:. Sharpe WF. The output of the RSI is a number on a scale from 0 to and it is typically calculated on a day basis. This a typical message book which contains the raw trading information. How to make money in commodities.
The dataset contained over 4. We focus on time series analysis, and more specifically, we examine features such as autocorrelation and partial autocorrelation, among others e. Accumulation Distribution Line. Procedia Computer Science. Multidimensional sequence classification based on fuzzy distances and discriminant analysis. Williams B. A SVM algorithm works on the given labeled data buy now with bitcoin insecure crypto exchanges hurting growth, and separates them via a boundary or a Hyperplane. Here, we adjust the trading rules according to the previous block horizon investigation to 5 and 34 previous MB blocks as follows:. We lag the indicator values to avoid look-ahead bias. More specifically, each of the features is evaluated based on the classification rate, the -norm of the predicted labels, and the ground truth. It is calculated as follows:. In this article, I show how to use a popular Python library for calculating TA indicators — TA-Lib — together with rsi scan macd moving average navigation trading mplied volatility indicator video zipline backtesting framework. Problem formulation HFT requires continuous analysis of market dynamics. High frequency trading using fuzzy momentum analysis. Chaikin oscillator. We should point out that we tested all the possible combinations for the five sorting methods and the three classifiers i.
About Help Legal. Blau W. We stop at this point, and in our next post on Machine learning we will see how framed rules like the ones devised above can be coded and backtested to check the viability of a trading strategy. The present work has identified a gap in the existing literature concerning the performance of technical indicators and comparisons with quantitative models. Time-series classification using neural bag-of-features. New concepts in technical trading systems. The number of top features that used in the above methods is different in each case and can be monitored in Figs 3 and 4. If you want to learn more about the basics of trading e. Average true range. The web app has a script that continuously updates the SQL database with new candles for each granularity. Using The Heikin-Ashi Technique. We calculate this indicator based on the closing prices of every MB block, as follows:. Smoothing and differentiation of data by simplified least squares procedures. Based on such a representation, a classifier is then employed to predict the state of the market at a time instance within a prediction horizon, as illustrated in Fig 1. This strategy can be considered an extension of the previous one — instead of a single moving average, we use two averages of different window sizes.
Heikin-Ashi [ 49 ] is a candlestick method and is described as a visual technique that eliminates irregularities:. Limit order books. Algorithmic trading uses computers, under specific rules, to rapidly perform accurate calculations based on statistical analysis. To the best of our knowledge this is the first attempt to compare these trading schools using several feature selection methods in a wrapper fashion in HFT. They wanted to trade every time two of these custom indicators intersected, and only at a certain angle. Some of the possible future directions:. SVM tries to maximize the margin around the separating hyperplane. The strategy we consider can be described as:. Fig 3. Technical analysis. Modelling financial time series. We see that the strategy generated much fewer signals than the one based on SMA. Bear in mind that many different window-lengths combinations defining the fast and slow MA can be considered for this strategy. Another aspect of quantitative analysis is building trading strategies such as mean-reversion as tested in [ 26 ].