Articles on data analysis and statistics in MQL5

This article focuses on specifics of choice, preconditioning and evaluation of the input variables (predictors) for use in machine learning models. New approaches and opportunities of deep predictor analysis and their influence on possible overfitting of models will be considered. The overall result of using models largely depends on the result of this stage. We will analyze two packages offering new and original approaches to the selection of predictors.

In this article the author talks about evolutionary calculations with the use of a personally developed genetic algorithm. He demonstrates the functioning of the algorithm, using examples, and provides practical recommendations for its usage.

In this article we will view seasonal characteristics of financial time series using Boxplot diagrams. Each separate boxplot (or box-and-whiskey diagram) provides a good visualization of how values are distributed along the dataset. Boxplots should not be confused with the candlestick charts, although they can be visually similar.

In this article, we start the development of the indicator buffer classes for the DoEasy library. We will create the base class of the abstract buffer which is to be used as a foundation for the development of different class types of indicator buffers.

In this article, we will consider creation of a symbol collection based on the abstract symbol object developed in the previous article. The abstract symbol descendants are to clarify a symbol data and define the availability of the basic symbol object properties in a program. Such symbol objects are to be distinguished by their affiliation with groups.

The article deals with storing data in the program's source code and creating audio and graphical files out of them. When developing an application, we often need audio and images. The MQL language features several methods of using such data.

While analyzing a huge number of trading strategies, orders for development of applications for MetaTrader 5 and MetaTrader 4 terminals and various MetaTrader websites, I came to the conclusion that all this diversity is based mostly on the same elementary functions, actions and values appearing regularly in different programs. This resulted in DoEasy cross-platform library for easy and quick development of МetaТrader 5 and МetaТrader 4 applications.

We continue the development of the functionality allowing users to trade using pending requests. In this article, we are going to implement the ability to place pending orders under certain conditions.

In the previous articles, we checked the concept of pending trading requests. A pending request is, in fact, a common trading order executed by a certain condition. In this article, we are going to create full-fledged classes of pending request objects — a base request object and its descendants.

Most of traders agree that the current market state analysis starts with the evaluation of higher chart timeframes. The analysis is performed downwards to lower timeframes until the one, at which deals are performed. This analysis method seems to be a mandatory part of professional approach for successful trading. In this article, we will discuss multi-timeframe indicators and their creation ways, as well as we will provide MQL5 code examples. In addition to the general evaluation of advantages and disadvantages, we will propose a new indicator approach using the MTF mode.

The major problem of applied statistics is the problem of accepting statistical hypotheses. It was long considered impossible to be solved. The situation has changed with the emergence of the eigen-coordinates method. It is a fine and powerful tool for a structural study of a signal allowing to see more than what is possible using methods of modern applied statistics. The article focuses on practical use of this method and sets forth programs in MQL5. It also deals with the problem of function identification using as an example the distribution introduced by Hilhorst and Schehr.

The article offers basic tools for the OLAP analysis of tester reports relating to single passes and optimization results. The tool can work with standard format files (tst and opt), and it also provides a graphical interface. MQL source codes are attached below.

This series of articles continues exploring deep neural networks (DNN), which are used in many application areas including trading. Here new dimensions of this theme will be explored along with testing of new methods and ideas using practical experiments. The first article of the series is dedicated to preparing data for DNN.

Before launching a robot on a trading account, we usually test and optimize it on quotes history. However, a reasonable question arises: how can past results help us in the future? The article describes applying the Monte Carlo method to construct custom criteria for trading strategy optimization. In addition, the EA stability criteria are considered.

In this article, we are going to store some data in the value of the orders and positions magic number and start the implementation of pending requests. To check the concept, let's create the first test pending request for opening market positions when receiving a server error requiring waiting and sending a repeated request.

In the previous articles, we started creating a large cross-platform library simplifying the development of programs for MetaTrader 5 and MetaTrader 4 platforms. In the fourth part, we tested tracking trading events on the account. In this article, we will develop trading event classes and place them to the event collections. From there, they will be sent to the base object of the Engine library and the control program chart.

The widget provides websites with a detailed release schedule of 500 indicators and indices, of the world's largest economies. Thus, traders quickly receive up-to-date information on all important events with explanations and graphs in addition to the main website content.

With the introduction of hedging, MetaTrader 5 provides an excellent opportunity to trade several Expert Advisors on a single trading account simultaneously. When one strategy is profitable, while the second one is loss-making, the profit graph may hang around zero. In this case, it is useful to build the Balance and Equity graphs for each trading strategy separately.

In the following article I am describing a process of implementing Moving Mini-Max indicator based on a paper by Z.G.Silagadze 'Moving Mini-max: a new indicator for technical analysis'. The idea of the indicator is based on simulation of quantum tunneling phenomena, proposed by G. Gamov in the theory of alpha decay.

This article is a continuation of the previous publication related to the creation of a graphical interface for optimization management. The article considers the logic of the add-on. A wrapper for the MetaTrader 5 terminal will be created: it will enable the running of the add-on as a managed process via C#. In addition, operation with configuration files and setup files is considered in this article. The application logic is divided into two parts: the first one describes the methods called after pressing a particular key, while the second part covers optimization launch and management.

Overbought/oversold zones characterize a certain state of the market, differentiating through weaker changes in the prices of securities. This adverse change in the synamics is pronounced most at the final stage in the development of trends of any scales. Since the profit value in trading depends directly on the capability of covering as large trend amplitude as possible, the accuracy of detecting such zones is a key task in trading with any securities whatsoever.

Starting with this article, we are going to develop a functionality allowing users to trade using pending requests under certain conditions, for example, when reaching a certain time limit, exceeding a specified profit or closing a position by stop loss.

In the upcoming series of articles, I will demonstrate the development of self-adapting algorithms considering most market factors, as well as show how to systematize these situations, describe them in logic and take them into account in your trading activity. I will start with a very simple algorithm that will gradually acquire theory and evolve into a very complex project.

In this article, we consider yet another custom trading strategy optimization criterion based on the balance graph analysis. The linear regression is calculated using the function from the ALGLIB library.

The article considers working with account events for tracking important changes in account properties affecting the automated trading. We have already implemented some functionality for tracking account events in the previous article when developing the account object collection.

In this article, we will start the development of the new library section - trading classes. Besides, we will consider the development of a unified base trading object for MetaTrader 5 and MetaTrader 4 platforms. When sending a request to the server, such a trading object implies that verified and correct trading request parameters are passed to it.

The article describes the basic principles and methods that allow you to analyze any strategy using spreadsheets (Excel, Calc, Google). The obtained results are compared with MetaTrader 5 tester.

The new version of the programming language for trading strategy development, MQL [MQL5], provides more powerful and effective features as compared with the previous version [MQL4]. The advantage essentially lies in the object-oriented programming features. This article looks into the possibility of using complex custom data types, such as nodes and lists. It also provides an example of using lists in practical programming in MQL5.

This article presents Pivot Mean Oscillator (PMO), an implementation of the cumulative moving average (CMA) as a trading indicator for the MetaTrader platforms. In particular, we first introduce Pivot Mean (PM) as a normalization index for timeseries that computes the fraction between any data point and the CMA. We then build PMO as the difference between the moving averages applied to two PM signals. Some preliminary experiments carried out on the EURUSD symbol to test the efficacy of the proposed indicator are also reported, leaving ample space for further considerations and improvements.

The article develops the ideas proposed in the previous part and considers them further. It describes the problems of yield distributions, plotting and studying statistical regularities.

This article deals with the theory and practical application of the algorithm for forecasting time series, based on support-vector method. It also proposes its implementation in MQL and provides test indicators and Expert Advisors. This technology has not been implemented in MQL yet. But first, we have to get to know math for it.

The article describes an attempt to combine theory with practice in the algorithmic trading field. Most of discussions concerning the creation of Trading Systems is connected with the use of historic bars and various indicators applied thereon. This is the most well covered field and thus we will not consider it. Bars represent a very artificial entity; therefore we will work with something closer to proto-data, namely the price ticks.

The article discusses the possibility of plotting statistical distribution histograms of market conditions with the help of the graphical memory meaning no indicator buffers and arrays are applied. Sample histograms are described in details and the "hidden" functionality of MQL5 graphical objects is shown.

In this article, we will analyze the concept of correlation between variables, as well as methods for the calculation of correlation coefficients and their practical use in trading. Correlation is a statistical relationship between two or more random variables (or quantities which can be considered random with some acceptable degree of accuracy). Changes in one ore more variables lead to systematic changes of other related variables.

The Strategy Tester in the MetaTrader 5 trading platform provides only two optimization options: complete search of parameters and genetic algorithm. This article proposes a new method for optimizing trading strategies — Simulated annealing. The method's algorithm, its implementation and integration into any Expert Advisor are considered. The developed algorithm is tested on the Moving Average EA.

In this article, we will create the class of a symbol object that is to be the basic object for creating the symbol collection. The class will allow us to obtain data on the necessary symbols for their further analysis and comparison.