Research Article
JPX Tokyo Stock Exchange Prediction with LightGBM
@INPROCEEDINGS{10.4108/eai.19-5-2023.2334371, author={Mingda Huo and Sen Wang and Tianxiao Xu and Daniel Boxiao Huang and Tong Zhou}, title={JPX Tokyo Stock Exchange Prediction with LightGBM}, proceedings={Proceedings of the 2nd International Conference on Bigdata Blockchain and Economy Management, ICBBEM 2023, May 19--21, 2023, Hangzhou, China}, publisher={EAI}, proceedings_a={ICBBEM}, year={2023}, month={7}, keywords={jpx tokyo stock exchange investment lightgbm sharpe ratio}, doi={10.4108/eai.19-5-2023.2334371} }- Mingda Huo
Sen Wang
Tianxiao Xu
Daniel Boxiao Huang
Tong Zhou
Year: 2023
JPX Tokyo Stock Exchange Prediction with LightGBM
ICBBEM
EAI
DOI: 10.4108/eai.19-5-2023.2334371
Abstract
The stock market is in an environment where risks and rewards coexist. A well-designed stock portfolio can make profits or even windfall profits through trading. Stock prediction refers to the use of scientific methods, using mathematical statistical methods as the means. In this paper, we pay attention to the JPX Tokyo Stock Exchange Prediction. The dataset is provided by Kaggle platform. We first do feature engineering and extract import features for model input. We use the LightGBM to predict the stock price. Sharpe Ratio is our evaluation metrics. The results show that our hybrid model owns the best performance with the highest Sharpe Ratio score 0.288, which is 0.023, 0.098, 0.048 higher than Xgboost, SVM and DNN respectively.
