The Comparison of LSTM, LGBM, and CNN in Stock Volatility Prediction

Abstract

In financial markets, volatility reflects the magnitude of price fluctuations. Forecasting volatility will be an important measure of the future direction of the market. Measuring and prdedicting stock market volatility has received increasing attention from academics and the industry over the past few years. This paper will focus on predicting the actual volatility of stocks using CNN, LightGBM, and LSTM models, using a data-set from Kaggle to make predictions. The paper gives a throughout analysis of the comparison for the performance of the three models. After testing with the chosen dataset, it was found that LGBM is more suitable for the task of predicting short-term stock volatility.