Forecasting Mutual Fund Prices on TEFAS Using LightGBM

Abstract

This paper investigates the application of the Light Gradient Boosting Machine (LightGBM) algorithm for predicting the prices of mutual funds traded on the Türkiye Electronic Fund Trading Platform (TEFAS). Given the increasing importance of mutual funds as an investment vehicle in Türkiye, this study explores the effectiveness of a state-of-the-art machine learning approach. Utilizing historical data from TEFAS, the LightGBM model is employed to capture complex patterns and non-linear relationships within the financial time series data. The research outlines the methodology for data preparation, feature implementation, data splitting, model configuration, training, and evaluation, including case studies to demonstrate the practical application and results

Keywords

• Machine learning
• LightGBM
• Tefas

References

1. Guennioui O, Chiadmi D, Amghar M. 2024. Global stock price forecasting during a period of market stress using LightGBM. Int J Comput Digit Syst, 15.1: 19-27.
2. Guo Y, Li Y, Xu Y. 2021. Study on the application of LSTM-LightGBM Model in stock rise and fall prediction. MATEC Web of Conferences, 336: 05011.
3. Hartanto A, Kholik YN, Pristyanto Y. 2023. Stock price time series data forecasting using the light gradient boosting machine (LightGBM) model. JOIV: Int J Inform Visualization, 7.4: 2270-2279.
4. Sun X, Liu M, Sima Z. 2020. A novel cryptocurrency price trend forecasting model based on LightGBM. Finance Res Lett, 32: 101084.
5. Tian L, Feng L, Yang L, Guo Y. 2022. Stock price prediction based on LSTM and LightGBM hybrid model. J. Supercomput, 78.9: 11768-11793.