StockTrendNet – Residual Learning-Based Deep Model for Accurate Stock Price Prediction

One of the main components of global finance is the stock market. Predicting future stock market prices with considerable accuracy can provide great insight into market movement and has the potential to generate profit. Among the many metrics used in the stock market, the closing price is one of the two most important, the other being the opening price. Despite such high importance, it remains difficult to predict the closing price, mainly due to the highly volatile nature of stock prices. To this end, we propose StockTrendNet, a custom Long Short-Term Memory (LSTM)-based model designed to capture intricate temporal patterns in stock market data, utilizing a gated memory mechanism with update and modulation gates, and incorporating residual and skip connections for enhanced robustness. StockTrendNet has been evaluated on publicly available financial datasets such as Nifty50, Sensex, S&P500, and Nikkei225, showcasing its superior predictive accuracy compared to existing models and positioning it as a robust framework for stock market forecasting. The code for our proposed model is available on https://github.com/Utathyaworks/StockTrendNet/