Stock Price Prediction of Turkish Banks Using Machine Learning Methods

Abstract

Stock markets are vital part of the economy since they utilize the capital. Investors in stock markets are providing capital with the expectation of positive return on their investment. In order to assess if an investment going to bring positive return, future prices of the stocks must be estimated. Thanks to development in technology and statistics, relatively more advanced estimation methods are developed and machine learning approaches are being integrated for this task. Following study is suggesting a stock price prediction model for Turkish banks using machine learning methods such as multiple linear regression, ridge regression, lasso regression, support vector machines, decision tree models, random forest, XGBoost method based on a wide dataset which is expanded using sliding windows method. After the models trained and tested, it has been observed that the XGBoost algorithm is superior to the other algorithms according to the result of the test errors. Thus the proposed model is convenient for predicting the stock prices of Turkish banks.

Decision Tree Analysis Approaches to Classify Sensors Data in a Water Pumping Station

“…The J48 algorithm can be used in a wide range of various fields, such as finance, health care, etc. For example, it is used in cardiology, stock identification, and data classification [36]. The J48 algorithm has many advantages over other algorithms.…”