Robust Deep Learning Models for Predicting the Trend of Stock Market Prices During Market Crash Periods

Investing in the stock market involves both opportunities and risks. However, the stock market may experience some fluctuations, which most investors regard as serious threats -- particularly when prices fall sharply due to external circumstances. At the beginning of 2020, as a consequence of the COVID-19 pandemic, equity market indices fell sharply. During this time, many investors suffered significant losses. Despite substantial research in stock market forecasting and the development of various efficient models, existing methods fall short in proposing sustainable and stable models during a financial crisis. To address this research gap, we propose two novel deep learning models: a Convolutional Neural Network (CNN)-based ensemble model, namely GAFECNN-Stacking and a stacking ensemble of enhanced WGANs-based model, namely GAFEWGAN, that maintains a high level of resilience to the stock market crash. The GAF-ECNN Stacking and GAF-EWGAN models achieved an average of 13.26% and 16.49% annual returns over 20 selected stocks.