Stock trading strategy based on regression learning algorithms
Subject Areas : Journal of Capital Market Analysis
NAASER HEYDARI
1
(Islamic Azad University, Arak, Iran)
مجید زنجیردار
2
(Department of Finance, Arak Branch, Islamic Azad University, Arak, Iran.)
Ali Lalbar
3
(Department of Finance, Arak Branch, Islamic Azad University, Arak, Iran.)
Keywords: Machine Learning, stock exchange, Trading Strategy, Regression Algorithms,
Abstract :
The aim of this study is to develop a stock trading strategy using regression learning algorithms. The researcher utilized the Yahoo Finance database to collect the necessary data using Python programming. Key technical analysis indicators and oscillators were calculated and incorporated into the model. The performance of the regression algorithms was evaluated using indicators such as determination coefficient, mean error of the mean, and square root of the error. Advanced statistical methods and software including Python, Spider, SPSS, and Excel were employed to analyze the differences between the evaluation indices of the designed algorithms. The Kruskal-Wallis test was used for meaningful comparison. Additionally, a diversified research sample consisting of companies from various sectors was chosen to generalize the findings. The selected companies were actively traded on the New York Stock Exchange with an average volume greater than 1 million and a market value larger than 200 trillion dollars. The sample was determined using a filter writing method on 28/06/2021 equal to 41 numbers as the sample of this research . The research was completed by the end of February 2023, and the random forest trading strategy model was identified as the most suitable approach.Keywords: Trading Strategy, Machine Learning, Regression Algorithms, Stock Exchange.
_||_