ENSEMBLE METHOD FOR THE PREDICTION OF STOCK PRICE USING MACHINE LEARNING TECHNIQUES

Authors: Author Name 1 (Affiliation 1), Author Name 2 (Affiliation 2)

Journal: Nigerian Journal Of Technical Education | Volume: 22 | Issue: 1 | Pages: 26 - 42

Keywords:Machine Learning, Ensemble Method, Support Vector Regressor, Convolutional Neural Network, Long Short- Term Memory, Stock Price Prediction

Abstract

An ensemble model, referred to as SVR-LSTM-CNN(Support Vector Regressor - Long Short- Term Memory - Convolutional Neural Network), was proposed for stock price forecasting . This model combines the SVR for capturing linear features and adeep neural network structure that incorporates LSTM and CNN layers to capture both linear and nonlinear data features. In the SVR-LSTM-CNN MODEL, SVR is utilised to capture linear features, LSTM captures long-term dependencies in the data, and CNN captures hierarchical data structures. The study focused on four different stocks from the Nigerian Stock Exchange historical database. A comprehensive performance evaluation, using daily stock prices, revealed that the SVR_LSTM_CNN model outperformed benchmark models in terms of Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE) performance measure.