Articles on machine learning in trading

In the previous article, we started exploring non-gradient optimization methods. We got acquainted with the genetic algorithm. Today, we will continue this topic and will consider another class of evolutionary algorithms.

Revolutionize your financial market analysis with Principal Component Analysis (PCA)! Discover how this powerful technique can unlock hidden patterns in your data, uncover latent market trends, and optimize your investment strategies. In this article, we explore how PCA can provide a new lens for analyzing complex financial data, revealing insights that would be missed by traditional approaches. Find out how applying PCA to financial market data can give you a competitive edge and help you stay ahead of the curve

GSA is a population optimization algorithm inspired by inanimate nature. Thanks to Newton's law of gravity implemented in the algorithm, the high reliability of modeling the interaction of physical bodies allows us to observe the enchanting dance of planetary systems and galactic clusters. In this article, I will consider one of the most interesting and original optimization algorithms. The simulator of the space objects movement is provided as well.

We continue our discussion of reinforcement learning algorithms for solving continuous action space problems. In this article, I will present the Soft Actor-Critic (SAC) algorithm. The main advantage of SAC is the ability to find optimal policies that not only maximize the expected reward, but also have maximum entropy (diversity) of actions.

Layered memory approaches that mimic human cognitive processes enable the processing of complex financial data and adaptation to new signals, thereby improving the effectiveness of investment decisions in dynamic markets.

K-Means clustering takes the approach to grouping data points as a process that’s initially focused on the macro view of a data set that uses random generated cluster centroids before zooming in and adjusting these centroids to accurately represent the data set. We will look at this and exploit a few of its use cases.

Join us in our discussion today as we look for an algorithmic procedure to minimize the total number of times we get stopped out of winning trades. The problem we faced is significantly challenging, and most solutions given in community discussions lack set and fixed rules. Our algorithmic approach to solving the problem increased the profitability of our trades and reduced our average loss per trade. However, there are further advancements to be made to completely filter out all trades that will be stopped out, our solution is a good first step for anyone to try.

This article outlines a simple strategy for normalizing the market data using the daily range and training a neural network to enhance market predictions. The developed models may be used in conjunction with an existing technical analysis frameworks or on a standalone basis to assist in predicting the overall market direction. The framework outlined in this article may be further refined by any technical analyst to develop models suitable for both manual and automated trading strategies.

Machine learning models are difficult to interpret and understanding why our models deviate from our expectations is critical if we want to gain any value from using such advanced techniques. Without comprehensive insight into the inner workings of our model, we might fail to spot bugs that are corrupting our model's performance, we may waste time over engineering features that aren't predictive and in the long run we risk underutilizing the power of these models. Fortunately, there is a sophisticated and well maintained all in one solution that allows us to see exactly what our model is doing underneath the hood.

Here we will describe only one of the aspects of machine learning - activation functions. In artificial neural networks, a neuron activation function calculates an output signal value based on the values of an input signal or a set of input signals. We will delve into the inner workings of the process.

Fish School Search (FSS) is a new optimization algorithm inspired by the behavior of fish in a school, most of which (up to 80%) swim in an organized community of relatives. It has been proven that fish aggregations play an important role in the efficiency of foraging and protection from predators.

In this article, we will create an arbitration system that remains legal in the eyes of brokers, creates thousands of synthetic prices on the Forex market, analyzes them, and successfully trades for profit.

In today’s discussion, we used alternative Daily data from the St. Louis Federal Reserve on the Broad US-Dollar Index and a collection of other macroeconomic indicators to predict the EURUSD future exchange rate. Unfortunately, while the data appears to have almost perfect correlation, we failed to realize any material gains in our model accuracy, possibly suggesting to us that investors may be better off using ordinary market quotes instead.

We continue to implement the approaches proposed by the authors of the FinCon framework. FinCon is a multi-agent system based on Large Language Models (LLMs). Today, we will implement the necessary modules and conduct comprehensive testing of the model on real historical data.

We have built a robust feature engineering pipeline using proper tick-based bars to eliminate data leakage and solved the critical problem of labeling with meta-labeled triple-barrier signals. This installment covers the advanced labeling technique, trend-scanning, for adaptive horizons. After covering the theory, an example shows how trend-scanning labels can be used with meta-labeling to improve on the classic moving average crossover strategy.

Discover the indispensable role of Support Vector Machines (SVM) in shaping the future of trading. This comprehensive guide explores how SVM can elevate your trading strategies, enhance decision-making, and unlock new opportunities in the financial markets. Dive into the world of SVM with real-world applications, step-by-step tutorials, and expert insights. Equip yourself with the essential tool that can help you navigate the complexities of modern trading. Elevate your trading game with SVM—a must-have for every trader's toolbox.

In this article series, I use experimentation and non-standard approaches to develop a profitable trading system and check whether neural networks can be of any help for traders. MetaTrader 5 is approached as a self-sufficient tool for using neural networks in trading.

We continue studying distributed Q-learning. Today we will look at this approach from the other side. We will consider the possibility of using quantile regression to solve price prediction tasks.

The article considers the Artificial Algae Algorithm (AAA) based on biological processes characteristic of microalgae. The algorithm includes spiral motion, evolutionary process and adaptation, which allows it to solve optimization problems. The article provides an in-depth analysis of the working principles of AAA and its potential in mathematical modeling, highlighting the connection between nature and algorithmic solutions.

In this article, we will rethink code locks, transforming them from security mechanisms into tools for solving complex optimization problems. Discover the world of code locks viewed not as simple security devices, but as inspiration for a new approach to optimization. We will create a whole population of "locks", where each lock represents a unique solution to the problem. We will then develop an algorithm that will "pick" these locks and find optimal solutions in a variety of areas, from machine learning to trading systems development.

The article explores the possibility of improving price forecasting based on trading volume analysis by integrating technical analysis principles with LSTM neural network architecture. Particular attention is paid to the detection and interpretation of anomalous volumes, the use of clustering and the creation of features based on volumes and their definition in the context of machine learning.

Data mining is crucial to a data scientist and a trader because very often, the data isn't as straightforward as we think it is. The human eye can not understand the minor underlying pattern and relationships in the dataset, maybe the K-means algorithm can help us with that. Let's find out...

The article presents an implementation of the Levenberg-Marquardt algorithm for training feedforward neural networks. A comparative analysis of performance with algorithms from the scikit-learn Python library has been conducted. Simpler learning methods, such as gradient descent, gradient descent with momentum, and stochastic gradient descent are preliminarily discussed.

Training Transformer models requires large amounts of data and is often difficult since the models are not good at generalizing to small datasets. The SAMformer framework helps solve this problem by avoiding poor local minima. This improves the efficiency of models even on limited training datasets.

The moving averages and the stochastic oscillator could be used to generate trend following trading signals. However, these signals will only be observed after the price action has occurred. We can effectively overcome this inherent lag in technical indicators using AI. This article will teach you how to create a fully autonomous AI-powered Expert Advisor in a manner that can improve any of your existing trading strategies. Even the oldest trading strategy possible can be improved.

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.

The amazing ability of weeds to survive in a wide variety of conditions has become the idea for a powerful optimization algorithm. IWO is one of the best algorithms among the previously reviewed ones.

In this article we discuss methods related to the analysis of timeseries in the frequency domain. Emphasizing the utility of examining the power spectra of time series when building predictive models. In this article we will discuss some of the useful perspectives to be gained by analyzing time series in the frequency domain using the discrete fourier transform (dft).

The study of the environment in reinforcement learning is a pressing problem. We have already looked at some approaches previously. In this article, we will have a look at yet another method based on maximizing the nuclear norm. It allows agents to identify environmental states with a high degree of novelty and diversity.

This article explores the development of an Expert Advisor (EA) for automated trading that combines technical analysis with deep learning predictions.

Whenever we consider reinforcement learning methods, we are faced with the issue of efficiently exploring the environment. Solving this issue often leads to complication of the algorithm and training of additional models. In this article, we will look at an alternative approach to solving this problem.

These advanced gradient-boosted decision tree techniques offer superior performance and flexibility, making them ideal for financial modeling and algorithmic trading. Learn how to leverage these tools to optimize your trading strategies, improve predictive accuracy, and gain a competitive edge in the financial markets.