Articles on machine learning in trading

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.

Nowadays, every trader must have heard of neural networks and knows how cool it is to use them. The majority believes that those who can deal with neural networks are some kind of superhuman. In this article, I will try to explain to you the neural network architecture, describe its applications and show examples of practical use.

Is it possible to develop an Expert Advisor able to optimize position open and close conditions at regular intervals according to the code commands? What happens if we implement a neural network (multilayer perceptron) in the form of a module to analyze history and provide strategy? We can make the EA optimize a neural network monthly (weekly, daily or hourly) and continue its work afterwards. Thus, we can develop a self-optimizing EA.

The article considers the possibility to apply Bayesian optimization to hyperparameters of deep neural networks, obtained by various training variants. The classification quality of a DNN with the optimal hyperparameters in different training variants is compared. Depth of effectiveness of the DNN optimal hyperparameters has been checked in forward tests. The possible directions for improving the classification quality have been determined.

Support Vector Machines have long been used in fields such as bioinformatics and applied mathematics to assess complex data sets and extract useful patterns that can be used to classify data. This article looks at what a support vector machine is, how they work and why they can be so useful in extracting complex patterns. We then investigate how they can be applied to the market and potentially used to advise on trades. Using the Support Vector Machine Learning Tool, the article provides worked examples that allow readers to experiment with their own trading.

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.

The article considers three methods which can be used to increase the classification quality of bagging ensembles, and their efficiency is estimated. The effects of optimization of the ELM neural network hyperparameters and postprocessing parameters are evaluated.

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.

This article considers new capabilities of the darch package (v.0.12.0). It contains a description of training of a deep neural networks with different data types, different structure and training sequence. Training results are included.

We continue to build ensembles. This time, the bagging ensemble created earlier will be supplemented with a trainable combiner — a deep neural network. One neural network combines the 7 best ensemble outputs after pruning. The second one takes all 500 outputs of the ensemble as input, prunes and combines them. The neural networks will be built using the keras/TensorFlow package for Python. The features of the package will be briefly considered. Testing will be performed and the classification quality of bagging and stacking ensembles will be compared.

This article considers using the Rattle package for automatic search of patterns for predicting long and short positions of currency pairs on Forex. This article can be useful both for novice and experienced traders.

The article discusses the methods for building and training ensembles of neural networks with bagging structure. It also determines the peculiarities of hyperparameter optimization for individual neural network classifiers that make up the ensemble. The quality of the optimized neural network obtained in the previous article of the series is compared with the quality of the created ensemble of neural networks. Possibilities of further improving the quality of the ensemble's classification are considered.

In previous articles, we used stochastic gradient descent to train a neural network using the same learning rate for all neurons within the network. In this article, I propose to look towards adaptive learning methods which enable changing of the learning rate for each neuron. We will also consider the pros and cons of this approach.

We continue studying the world of neural networks. In this article, we will consider another type of neural networks, recurrent networks. This type is proposed for use with time series, which are represented in the MetaTrader 5 trading platform by price charts.

In this second article, we will continue to study neural networks and will consider an example of using our created CNet class in Expert Advisors. We will work with two neural network models, which show similar results both in terms of training time and prediction accuracy.

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 second article of the series about deep neural networks will consider the transformation and choice of predictors during the process of preparing data for training a model.

Perhaps one of the most advanced models among currently existing language neural networks is GPT-3, the maximal variant of which contains 175 billion parameters. Of course, we are not going to create such a monster on our home PCs. However, we can view which architectural solutions can be used in our work and how we can benefit from them.

In addition to creation of neuronets, the NeuroSolutions software suite allows exporting them as DLLs. This article describes the process of creating a neuronet, generating a DLL and connecting it to an Expert Advisor for trading in MetaTrader 5.

As a continuation of the neural network topic, I propose considering convolutional neural networks. This type of neural network are usually applied to analyzing visual imagery. In this article, we will consider the application of these networks in the financial markets.