Research Article
Research on stock prediction based on LSTM and CatBoost algorithm
@INPROCEEDINGS{10.4108/eai.19-5-2023.2334326, author={Yu Sun and Liwei Tian}, title={Research on stock prediction based on LSTM and CatBoost algorithm}, 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={lstm catboost bayesian optimization stock price forecasting time series data}, doi={10.4108/eai.19-5-2023.2334326} }
- Yu Sun
Liwei Tian
Year: 2023
Research on stock prediction based on LSTM and CatBoost algorithm
ICBBEM
EAI
DOI: 10.4108/eai.19-5-2023.2334326
Abstract
Stock prediction is a classical problem at the intersection of computer science and finance. How to find an accurate, stable and effective model to predict the rise and fall of stocks has become a hot research topic among financial scholars. In the face of the increasingly prominent demand for stock analysis technology, combined forecasting model began to develop and achieved a lot of results. In this paper, we take the future financial time series up-down trend as the forecast goal, take the stock history data attribute value as the research object, based on the depth machine learning method, the combination model of LSTM and CatBoost optimized by Bayesian algorithm is used to predict the rise and fall of stocks. The model is validated by three evaluation indexes: MSE, MAE and Accuracy, it is concluded that the LSTM-BO-CatBoost model is more stable and feasible than LSTM-CatBoost, LSTM-XGBoost hybrid model, single LSTM network model and RNN network model.