Short-Term Stock Market Price Trend Prediction Using Genetic Algorithm-Enhanced Back Propagation Neural Networks for Traders’ Sustainable and Inclusive Future

This study focuses on predicting short-term stock prices in the complex and unpredictable stock market environment. The proposed model integrates a hybrid approach, combining Genetic Algorithm (GA) as a metaheuristic algorithm and Back Propagation Neural Networks (BPNN) as a neural network model. The evaluation utilizes datasets from Infosys and Hindustan Unilever Limited stock markets and involves preprocessing steps such as normalization and wavelet transformation, along with the application of stock technical indicators. GA is employed for dimensionality reduction, and the BPNN model is trained and tested for prediction analysis. Model performance is assessed using metrics such as MAE, RMSE, and MAPE, compared to other models (ANN, BPNN, RNN) for validation on both datasets. The comprehensive 20-day evaluation underscores the model's efficacy and reliability, making it an appealing option for financial forecasting in dynamic market conditions. This provides valuable insights for investors and financial analysts.