Neural networks made easy (Part 27): Deep Q-Learning (DQN)
We continue to study reinforcement learning. In this article, we will get acquainted with the Deep Q-Learning method. The use of this method has enabled the DeepMind team to create a model that can outperform a human when playing Atari computer games. I think it will be useful to evaluate the possibilities of the technology for solving trading problems.
Neural networks made easy (Part 26): Reinforcement Learning
We continue to study machine learning methods. With this article, we begin another big topic, Reinforcement Learning. This approach allows the models to set up certain strategies for solving the problems. We can expect that this property of reinforcement learning will open up new horizons for building trading strategies.
Neural networks made easy (Part 25): Practicing Transfer Learning
In the last two articles, we developed a tool for creating and editing neural network models. Now it is time to evaluate the potential use of Transfer Learning technology using practical examples.
Market math: profit, loss and costs
In this article, I will show you how to calculate the total profit or loss of any trade, including commission and swap. I will provide the most accurate mathematical model and use it to write the code and compare it with the standard. Besides, I will also try to get on the inside of the main MQL5 function to calculate profit and get to the bottom of all the necessary values from the specification.
Population optimization algorithms
This is an introductory article on optimization algorithm (OA) classification. The article attempts to create a test stand (a set of functions), which is to be used for comparing OAs and, perhaps, identifying the most universal algorithm out of all widely known ones.
Neural networks made easy (Part 22): Unsupervised learning of recurrent models
We continue to study unsupervised learning algorithms. This time I suggest that we discuss the features of autoencoders when applied to recurrent model training.
Neural networks made easy (Part 21): Variational autoencoders (VAE)
In the last article, we got acquainted with the Autoencoder algorithm. Like any other algorithm, it has its advantages and disadvantages. In its original implementation, the autoenctoder is used to separate the objects from the training sample as much as possible. This time we will talk about how to deal with some of its disadvantages.
Data Science and Machine Learning (Part 07): Polynomial Regression
Unlike linear regression, polynomial regression is a flexible model aimed to perform better at tasks the linear regression model could not handle, Let's find out how to make polynomial models in MQL5 and make something positive out of it.
Risk and capital management using Expert Advisors
This article is about what you can not see in a backtest report, what you should expect using automated trading software, how to manage your money if you are using expert advisors, and how to cover a significant loss to remain in the trading activity when you are using automated procedures.
Neural networks made easy (Part 20): Autoencoders
We continue to study unsupervised learning algorithms. Some readers might have questions regarding the relevance of recent publications to the topic of neural networks. In this new article, we get back to studying neural networks.
Matrix and Vector operations in MQL5
Matrices and vectors have been introduced in MQL5 for efficient operations with mathematical solutions. The new types offer built-in methods for creating concise and understandable code that is close to mathematical notation. Arrays provide extensive capabilities, but there are many cases in which matrices are much more efficient.
The price movement model and its main provisions (Part 2): Probabilistic price field evolution equation and the occurrence of the observed random walk
The article considers the probabilistic price field evolution equation and the upcoming price spike criterion. It also reveals the essence of price values on charts and the mechanism for the occurrence of a random walk of these values.
Neural networks made easy (Part 19): Association rules using MQL5
We continue considering association rules. In the previous article, we have discussed theoretical aspect of this type of problem. In this article, I will show the implementation of the FP Growth method using MQL5. We will also test the implemented solution using real data.
Neural networks made easy (Part 18): Association rules
As a continuation of this series of articles, let's consider another type of problems within unsupervised learning methods: mining association rules. This problem type was first used in retail, namely supermarkets, to analyze market baskets. In this article, we will talk about the applicability of such algorithms in trading.
Data Science and Machine Learning — Neural Network (Part 02): Feed forward NN Architectures Design
There are minor things to cover on the feed-forward neural network before we are through, the design being one of them. Let's see how we can build and design a flexible neural network to our inputs, the number of hidden layers, and the nodes for each of the network.
Metamodels in machine learning and trading: Original timing of trading orders
Metamodels in machine learning: Auto creation of trading systems with little or no human intervention — The model decides when and how to trade on its own.
Data Science and Machine Learning — Neural Network (Part 01): Feed Forward Neural Network demystified
Many people love them but a few understand the whole operations behind Neural Networks. In this article I will try to explain everything that goes behind closed doors of a feed-forward multi-layer perception in plain English.
Neural networks made easy (Part 17): Dimensionality reduction
In this part we continue discussing Artificial Intelligence models. Namely, we study unsupervised learning algorithms. We have already discussed one of the clustering algorithms. In this article, I am sharing a variant of solving problems related to dimensionality reduction.
Developing a trading Expert Advisor from scratch (Part 17): Accessing data on the web (III)
In this article we continue considering how to obtain data from the web and to use it in an Expert Advisor. This time we will proceed to developing an alternative system.
Data Science and Machine Learning (Part 06): Gradient Descent
The gradient descent plays a significant role in training neural networks and many machine learning algorithms. It is a quick and intelligent algorithm despite its impressive work it is still misunderstood by a lot of data scientists let's see what it is all about.
Neural networks made easy (Part 16): Practical use of clustering
In the previous article, we have created a class for data clustering. In this article, I want to share variants of the possible application of obtained results in solving practical trading tasks.
Developing a trading Expert Advisor from scratch (Part 16): Accessing data on the web (II)
Knowing how to input data from the Web into an Expert Advisor is not so obvious. It is not so easy to do without understanding all the possibilities offered by MetaTrader 5.
The price movement model and its main provisions (Part 1): The simplest model version and its applications
The article provides the foundations of a mathematically rigorous price movement and market functioning theory. Up to the present, we have not had any mathematically rigorous price movement theory. Instead, we have had to deal with experience-based assumptions stating that the price moves in a certain way after a certain pattern. Of course, these assumptions have been supported neither by statistics, nor by theory.
Neural networks made easy (Part 15): Data clustering using MQL5
We continue to consider the clustering method. In this article, we will create a new CKmeans class to implement one of the most common k-means clustering methods. During tests, the model managed to identify about 500 patterns.
Developing a trading Expert Advisor from scratch (Part 15): Accessing data on the web (I)
How to access online data via MetaTrader 5? There are a lot of websites and places on the web, featuring a huge amount information. What you need to know is where to look and how best to use this information.
Neural networks made easy (Part 14): Data clustering
It has been more than a year since I published my last article. This is quite a lot time to revise ideas and to develop new approaches. In the new article, I would like to divert from the previously used supervised learning method. This time we will dip into unsupervised learning algorithms. In particular, we will consider one of the clustering algorithms—k-means.
Data Science and Machine Learning (Part 05): Decision Trees
Decision trees imitate the way humans think to classify data. Let's see how to build trees and use them to classify and predict some data. The main goal of the decision trees algorithm is to separate the data with impurity and into pure or close to nodes.
MQL5 Wizard techniques you should know (Part 01): Regression Analysis
Todays trader is a philomath who is almost always (either consciously or not...) looking up new ideas, trying them out, choosing to modify them or discard them; an exploratory process that should cost a fair amount of diligence. This clearly places a premium on the trader's time and the need to avoid mistakes. These series of articles will proposition that the MQL5 wizard should be a mainstay for traders. Why? Because not only does the trader save time by assembling his new ideas with the MQL5 wizard, and greatly reduce mistakes from duplicate coding; he is ultimately set-up to channel his energy on the few critical areas of his trading philosophy.
Data Science and Machine Learning (Part 04): Predicting Current Stock Market Crash
In this article I am going to attempt to use our logistic model to predict the stock market crash based upon the fundamentals of the US economy, the NETFLIX and APPLE are the stocks we are going to focus on, Using the previous market crashes of 2019 and 2020 let's see how our model will perform in the current dooms and glooms.
Data Science and Machine Learning (Part 03): Matrix Regressions
This time our models are being made by matrices, which allows flexibility while it allows us to make powerful models that can handle not only five independent variables but also many variables as long as we stay within the calculations limits of a computer, this article is going to be an interesting read, that's for sure.
Tips from a professional programmer (Part III): Logging. Connecting to the Seq log collection and analysis system
Implementation of the Logger class for unifying and structuring messages which are printed to the Experts log. Connection to the Seq log collection and analysis system. Monitoring log messages online.
Data Science and Machine Learning (Part 02): Logistic Regression
Data Classification is a crucial thing for an algo trader and a programmer. In this article, we are going to focus on one of classification logistic algorithms that can probability help us identify the Yes's or No's, the Ups and Downs, Buys and Sells.
Mathematics in trading: Sharpe and Sortino ratios
Return on investments is the most obvious indicator which investors and novice traders use for the analysis of trading efficiency. Professional traders use more reliable tools to analyze strategies, such as Sharpe and Sortino ratios, among others.
Data Science and Machine Learning (Part 01): Linear Regression
It's time for us as traders to train our systems and ourselves to make decisions based on what number says. Not on our eyes, and what our guts make us believe, this is where the world is heading so, let us move perpendicular to the direction of the wave.
The correct way to choose an Expert Advisor from the Market
In this article, we will consider some of the essential points you should pay attention to when purchasing an Expert Advisor. We will also look for ways to increase profit, to spend money wisely, and to earn from this spending. Also, after reading the article, you will see that it is possible to earn even using simple and free products.
Visual evaluation of optimization results
In this article, we will consider how to build graphs of all optimization passes and to select the optimal custom criterion. We will also see how to create a desired solution with little MQL5 knowledge, using the articles published on the website and forum comments.
An Analysis of Why Expert Advisors Fail
This article presents an analysis of currency data to better understand why expert advisors can have good performance in some regions of time and poor performance in other regions of time.
Combinatorics and probability for trading (Part V): Curve analysis
In this article, I decided to conduct a study related to the possibility of reducing multiple states to double-state systems. The main purpose of the article is to analyze and to come to useful conclusions that may help in the further development of scalable trading algorithms based on the probability theory. Of course, this topic involves mathematics. However, given the experience of previous articles, I see that generalized information is more useful than details.
Combinatorics and probability for trading (Part IV): Bernoulli Logic
In this article, I decided to highlight the well-known Bernoulli scheme and to show how it can be used to describe trading-related data arrays. All this will then be used to create a self-adapting trading system. We will also look for a more generic algorithm, a special case of which is the Bernoulli formula, and will find an application for it.
Multilayer perceptron and backpropagation algorithm (Part II): Implementation in Python and integration with MQL5
There is a Python package available for developing integrations with MQL, which enables a plethora of opportunities such as data exploration, creation and use of machine learning models. The built in Python integration in MQL5 enables the creation of various solutions, from simple linear regression to deep learning models. Let's take a look at how to set up and prepare a development environment and how to use use some of the machine learning libraries.