StockPred: a framework for stock Price prediction

Abstract

Recently, Stock Price prediction becomes a significant practical aspect of the economic arena. The stock price prediction is generally considered as one of the most exciting challenges due to the noise and volatility characteristics of stock market behavior. Therefore, this paper proposes a framework to address these challenges and efficiently predicting stock price using learning models such as Long Short Term Memory (LSTM), Convolutional Neural Network (CNN), Support Vector Machine (SVM), Linear Regression, Logistic Regression, K-Neighbors, Decision Tree, Random Forest, Stacked-LSTM, and Bidirectional-LSTM. Numerous experiments with different scenarios are performed to evaluate the projected framework with the stock price dataset. The results demonstrate that the applied models within the framework such as the CNN model outperformed the other models in stock price prediction at different circumstances based on several evaluation metrics like R-Square (R2), Root Mean Square Error (RMSE), Root Mean Square (RMS), Mean Square Error (MSE), Mean Average Error (MAE) and Mean Average Percentage Error (MAPE).