An enhanced prediction model for Nigerian stock exchange market

Abstract

This study aims to enhance a predictive model for Nigerian stock exchange market (NSEM). To develop a prediction model, Back propagation network (BPN) with Levenberg-Marquardt and scaled conjugate gradient descent algorithms were implemented. The same transformed dataset of the NSEM used by existing method was employed in this study. Eventually, artificial neural network with Levenberg-Marquardt approach (ANN-LM) was found to be more competent in predicting NSE market with mean square error (MSE) and regression, R as 0.0049645 and 0.9936, respectively. This paper starts with a brief introduction of NSE, artificial neural network (ANN) and machine learning (ML) models used for prediction. System
design and data normalization using MINITAB software were described. Training algorithm, performance of a predictive model and graphs response of output and target were analyzed. The software used for the implementation is MATLAB R2014a (version 8.3) and executed in PC Intel Pentium IV E7400 processor with 2.80 GHz speed and 2.0 GB of RAM.