A Neuro-Genetic Technique for Index Prediction

Artificial Neural Network (ANN) has preeminent learning ability, but often exhibit inconsistent and unpredictable performance for noisy data. In addition, it may not be possible to train ANN or the training task cannot be effectively carried out without data reduction when the amount of data is so large. In this paper, we have used a Neuro-genetic model to predict the index value for Stock Price Index of Bombay Stock Exchange. Here we used the genetic algorithm (GA) to optimize simultaneously the connection weights between layers and a selection task for relevant parameters such as Bias to the hidden as well as output layer. The globally evolved weights mitigate the well-known limitations of gradient descent algorithm. In addition, genetically selected weights and parameters shorten the learning time and enhance prediction performance. Experimental results show that the Neuro-genetic approach is a promising method for Index Prediction than ANN model. Index Terms : Genetic algorithms; artificial neural networks; Index Prediction, Bombay Stock Exchange.