Artificial Neural Network Models for Forecasting Stock Price Index in the Bombay Stock Exchange

Abstract

Artificial Neural Network (ANN) has been shown to be an efficient tool for non-parametric modelling of data in a variety of different contexts where the output is a non-linear function of the inputs. These include business forecasting, credit scoring, bond rating, business failure prediction, medicine, pattern recognition and image processing. A large number of studies have been reported in literature with reference to the use of ANN in modelling stock prices in western countries. However, not much work along these lines has been reported in the Indian context. In this article we discuss the modelling of the Indian stock market (price index) data using ANN. We study the efficacy of ANN in modelling the Bombay Stock Exchange (BSE) SENSEX weekly closing values. We develop two networks with three hidden layers for the purpose of this study which are denoted as ANN1 and ANN2. ANN1 takes as its inputs the weekly closing value, 52-week moving average of the weekly closing SENSEX values, 5-week moving average of the same, and the 10-week Oscillator for the past 200 weeks. ANN2 takes as its inputs the weekly closing value, 52-week moving average of the weekly closing SENSEX values, 5-week moving average of the same and the 5-week volatility for the past 200 weeks. Both the neural networks are trained using data for 250 weeks starting January 1997. To assess the performance of the networks we used them to predict the weekly closing SENSEX values for the two-year period beginning January 2002. The root mean square error (RMSE) and mean absolute error (MAE) are chosen as indicators of performance of the networks. ANN1 achieved an RMSE of 4.82 per cent and MAE of 3.93 per cent while ANN2 achieved an RMSE of 6.87 per cent and MAE of 5.52 per cent.

Predicting stock market trends using hybrid ant-colony-based data mining algorithms: an empirical validation on the Bombay Stock Exchange

“…Ravichandran et al (2005) used a feed-forward back propagation network to predict the stocks trading on the Bombay Stock Exchange (BSE). Dutta et al (2006) used feed-forward ANN models for forecasting the BSE sensitive index (BSE-SENSEX) with reasonable accuracy. Swanson and White (1997) suggested that a single hidden layer feed-forward ANN offers a useful and flexible alternative to fixed specification linear models.…”

Predicting Next Trading Day Closing Price of Qatar Exchange Index Using Technical Indicators and Artificial Neural Networks

“…The key element of this paradigm is the novel structure of the information processing system (Haykin, ); that is, how layers of neurons (also called processing elements) are organized and connected. The three essential features of an ANN structure are (1) the basic computing elements, referred to as neurons, (2) the network architecture describing the connections between the neurons and (3) the training algorithm used to find values of the network parameters for performing a particular task (Dutta et al , ). The neurons that receive input represent the input layer, and the neurons that generate the output represent the output layer.…”

Multidimensional Liquidity: Evidences from Indian Stock Market

“…However, machine learning algorithms were used by some authors for stock market prediction. While Dutta et al (2006) evidence that ANN performs satisfactorily in predicting closing prices of SENSEX, the leading index of Bombay Stock Exchange, Qin et al (2012) evidence support for Random Forest based trading model for the Singapore exchange. Sala (2011) develops an alternative approach of liquidity risk modelling using a recurrent neural network and shows that machine learning may be an important alternative while modelling liquidity risk.…”