Performance Comparison of Deep Learning Models in Predicting Stock Prices

The stock market in India is highly volatile, dynamic, and nonlinear, presenting significant challenges for researchers in time series forecasting. Predicting stock prices accurately has remained a critical topic of interest for both financial analysts and technical researchers. In this study, we deploy four deep learning models—LSTM, GRU, Simple RNN, and CNN1D—to predict the one-day-ahead closing prices of 20 prominent Indian companies grouped into five sectors: Communication Services, Consumer Cyclical, Energy, Healthcare, and Technology. The models were evaluated using five metrics: Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), Mean Absolute Percentage Error (MAPE), Normalized RMSE (NRMSE), and Coefficient of Determination (R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>). Our experimental results provide valuable insights into the comparative performance of these models in predicting stock prices, contributing to a deeper understanding of their applicability in Indian financial markets.