Asian Journal of Research in Banking and Finance
Year : 2011, Volume : 1, Issue : 1
First page : ( 1) Last page : ( 19)
Online ISSN : 2249–7323.

An empirical study in indian share market using neural network and genetic algorithm

Ghoshal Arnab Kumar*, Mukherjee Tuhin*, Dhar Satyajit**

*Department of Computer Science and Application, Ramakrishna Mission Vidyamandira, India.

**Department of Business Administration, University of Kalyani, India.

Online published on 1 November, 2011.

Abstract

This paper performs an experiment to forecast stock market movement in India using Artificial Neural Network (ANN) and Genetic Algorithm (GA). This model is named Genetically optimized Neural Network (GNN). We have tested this newly created model against traditional ARCH/GARCH models using z-test.We have used different error metrics like Average Absolute Error (AAE), Mean Absolute Percentage Error(MAPE), Mean Square Error(MSE), Max AE, R-SQ during our comparative study. This paper concludes the difference of predictive ability of our model with that of traditional ARCH/GARCH models.

Top

Keywords

Artificial Neural Network (ANN) forecasting models, GeneticAlgorithm(GA).

Top