Unveling the Precision of Deep Learning Models for Stock Price Prediction: A Comparative Analysis of Bi-LSTM, LSTM, and GRU

Due to the unpredictability of the stock market, accurate prognostic models are necessary for investing. In recent years, machine learning techniques, specifically deep learning algorithms, have grown in popularity for predicting stock prices. This paper seeks to compare the stock-price forecasting abilities of several deep learning models, including LSTM, Bi-LSTM, and GRU. The algorithms make use of the capabilities of Recurrent Neural Networks (RNNs), with a particular emphasis on the Long-Short Term Memory (LSTM) model. The primary objective is to evaluate the accuracy of these machine learning algorithms at predicting stock market values and to determine how the number of training epochs affects model performance. Through comparative analysis, we intend to identify the most accurate model for predicting stock prices. Using historical stock market data, the research involves training and evaluating the various models. Common evaluation metrics, such as Root Mean Square Error (RMSE), Mean Squared Error (MSE), and Mean Absolute Error (MAE), are used to evaluate the performance of each model. In terms of RMSE, MSE, and MAE, the bi-LSTM model outperforms the other models, obtaining values of 0.00029, 0.00029, and 0.01 respectively.