Comparing the approach of machine learning algorithms in predicting the maintenance costs of trading strategies

Document Type : Research Paper

Authors

1 Assistant Professor, Department of Accounting, Islamic Azad University, Arak, Iran (corresponding author)

2 Master of Accounting, Islamic Azad University

3 Assistant Professor of Accounting Department, Islamic Azad University, Arak

4 PhD student, financial engineering, Islamic Azad University, Arak, Iran

10.22051/jfm.2024.46462.2902

Abstract

The purpose of this research was to compare the approach of machine learning algorithms in predicting the maintenance costs of trading strategies. The current research, in terms of its practical purpose and based on the method of data collection, is part of the non-experimental research method and examines the relationships between the variables and describes the variables and finally presents a model that is based on the inductive method. It will be generalizable to the entire statistical population; In this research, the library method was used to collect data and information, and by collecting data of sample companies by referring to financial statements, explanatory notes and the stock exchange monthly. accepted Based on the systematic elimination method, 150 companies were selected as a statistical sample; In order to describe and summarize the collected data, descriptive and inferential statistics have been used. In order to analyze the data, non-linear methods of decision trees and neural networks were used, and Excel 2016, Weka9 and Matlab 2019 software were used to confirm and reject the research hypotheses. The results showed that machine learning algorithms such as decision tree algorithm and neural networks have a high ability to predict the maintenance costs of trading strategies; In addition, the ability of the aforementioned algorithms, i.e. decision tree algorithm and artificial neural networks to predict the maintenance costs of trading strategies are statistically the same and no significant difference was found between them.

Keywords

Main Subjects

Financial Management Strategy

Articles in Press, Accepted Manuscript
Available Online from 10 September 2024

Files

History

  • Receive Date: 18 February 2024
  • Revise Date: 10 July 2024
  • Accept Date: 10 September 2024

Share

How to cite

Statistics