Chao Jie Shen / Profile

Finding rules for a trade system and programming them in an Expert Advisor is a half of the job. Somehow, you need to correct the operation of the Expert Advisor as it accumulates the results of trading. This article describes one of approaches, which allows improving performance of an Expert Advisor through creation of a feedback that measures slope of the balance curve.

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.

If specific neural network programs for trading seem expensive and complex or, on the contrary, too simple, try NeuroPro. It is free and contains the optimal set of functionalities for amateurs. This article will tell you how to use it in conjunction with MetaTrader 5.

The market price is formed out of a stable balance between demand and supply which, in turn, depend on a variety of economic, political and psychological factors. Differences in nature as well as causes of influence of these factors make it difficult to directly consider all the components. This article sets forth an attempt to predict the market price on the basis of an elaborated regression model.

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.

Genetic (evolutionary) algorithms are used for optimization purposes. An example of such purpose can be neuronet learning, i.e., selection of such weight values that allow reaching the minimum error. At this, the genetic algorithm is based on the random search method.

This article considers the application of multiple regression analysis to macroeconomic statistics. It also gives an insight into the evaluation of the statistics impact on the currency exchange rate fluctuation based on the example of the currency pair EURUSD. Such evaluation allows automating the fundamental analysis which becomes available to even novice traders.

We now know that probability density function (PDF) of a market cycle does not remind a Gaussian but rather a PDF of a sine wave and most of the indicators assume that the market cycle PDF is Gaussian we need a way to "correct" that. The solution is to use Fisher Transform. The Fisher transform changes PDF of any waveform to approximately Gaussian. This article describes the mathematics behind the Fisher Transform and the Inverse Fisher Transform and their application to trading. A proprietary trading signal module based on the Inverse Fisher Transform is presented and evaluated.

The article addresses probability distributions (normal, log-normal, binomial, logistic, exponential, Cauchy distribution, Student's t-distribution, Laplace distribution, Poisson distribution, Hyperbolic Secant distribution, Beta and Gamma distribution) of random variables used in Applied Statistics. It also features classes for handling these distributions.

In this article, we will develop a tool for CFTC report analysis. We will solve the following problem: to develop an indicator, that allows using the CFTC report data directly from the data files provided by Commission without an intermediate processing and conversion. Further, it can be used for the different purposes: to plot the data as an indicator, to proceed with the data in the other indicators, in the scripts for the automated analysis, in the Expert Advisors for the use in the trading strategies.

Estimation of statistical parameters of a sequence is very important, since most of mathematical models and methods are based on different assumptions. For example, normality of distribution law or dispersion value, or other parameters. Thus, when analyzing and forecasting of time series we need a simple and convenient tool that allows quickly and clearly estimating the main statistical parameters. The article shortly describes the simplest statistical parameters of a random sequence and several methods of its visual analysis. It offers the implementation of these methods in MQL5 and the methods of visualization of the result of calculations using the Gnuplot application.

There are a lot of measures that allow determining the effectiveness and profitability of a trade system. However, traders are always ready to put any system to a new crash test. The article tells how the statistics based on measures of effectiveness can be used for the MetaTrader 5 platform. It includes the class for transformation of the interpretation of statistics by deals to the one that doesn't contradict the description given in the "Statistika dlya traderov" ("Statistics for Traders") book by S.V. Bulashev. It also includes an example of custom function for optimization.

How many cores do you have on your home computer? How many computers can you use to optimize a trading strategy? We show here how to use the MQL5 Cloud Network to accelerate calculations by receiving the computing power across the globe with the click of a mouse. The phrase "Time is money" becomes even more topical with each passing year, and we cannot afford to wait for important computations for tens of hours or even days.

MetaTrader 5 allows us to simulate automatic trading, within an embedded strategy tester, by using Expert Advisors and the MQL5 language. This type of simulation is called testing of Expert Advisors, and can be implemented using multithreaded optimization, as well as simultaneously on a number of instruments. In order to provide a thorough testing, a generation of ticks based on the available minute history, needs to be performed. This article provides a detailed description of the algorithm, by which the ticks are generated for the historical testing in the MetaTrader 5 client terminal.

This article is dedicated to a new and perspective direction in machine learning - deep learning or, to be precise, deep neural networks. This is a brief review of second generation neural networks, the architecture of their connections and main types, methods and rules of learning and their main disadvantages followed by the history of the third generation neural network development, their main types, peculiarities and training methods. Conducted are practical experiments on building and training a deep neural network initiated by the weights of a stacked autoencoder with real data. All the stages from selecting input data to metric derivation are discussed in detail. The last part of the article contains a software implementation of a deep neural network in an Expert Advisor with a built-in indicator based on MQL4/R.

What are the differences between the three modes of testing in MetaTrader 5, and what should be particularly looked for? How does the testing of an EA, trading simultaneously on multiple instruments, take place? When and how are the indicator values calculated during testing, and how are the events handled? How to synchronize the bars from different instruments during testing in an "open prices only" mode? This article aims to provide answers to these and many other questions.

The article compares the time and results of Expert Advisors' optimization using genetic algorithms and those obtained by simple search.

Step-by-step instructions of how to organize data transfer from Matlab to MetaTrader 4 using DDE.

The article describes the methods of how to understand the tester optimization results better. It also gives some tips that help to avoid "harmful optimization".