Comparative Analysis Artificial Neural Network's Back Propogation Algorithm to statistical Least Squre Method In Security Prediction Using Nigerian Stock Exchange Market

Abstract

Statistical analysis has often been used in predicting financial operations in Nigerian economy. In this work. artificial neural network was used to predict movements in stock prices in Nigerian Stock Exchange market. Studies were carried out for the prediction of stock index values as well as daily direction of changes in the index. A network was designed using Back Propagation Algorithm (BPA) to predict stock index values and prices in the exchange for a period of 90 days. The data collected during this period was processed using the BPA algorithm to get an output such that the error between the actual indices and prices, and the computed output was brought to minimum. About 90% of the data was used for the actual training while the remaining 10% was used as test data. The same data was also processed using the Least Squares (LS) method. The results show that BPA algorithm has superior performances in terms of the accuracy of prediction over the LS method. ihis result of the study is useful to stock market operators.