Title: Modelling and forecasting volatility for BSE and NSE stock index: linear vs. nonlinear approach
Authors: A. Shanthi; R. Thamilselvan
Addresses: School of Management Studies, Sathyabama Institute of Science and Technology, Chennai, 600101, India ' School of Management Studies, Sathyabama Institute of Science and Technology, Chennai, 600101, India
Abstract: This article addresses the important issues related to stock market volatility by modelling and forecasting in the Sensitivity Index of Bombay Stock Exchange and Nifty 50 of National Stock Exchange in India by employing linear models and nonlinear models to show the usefulness of the model by attempting the sample forecast from 1 January 1996 to 31 December 2004, 1 January 2006 to 31 December 2014 and 1 January 1996 to 31 December 2014 as an in-sample forecast, by considering pre-period, post-period and full period. Apart from that, the out-of-sample forecast is also attempted in this study to draw a valid conclusion by using forecasting models. Overall, the study suggests that volatility is a part and parcel of the stock market, which is influenced by other key determining factors such as like inflow of foreign capital into the country, exchange rate, balance of payment and interest rate.
Keywords: modelling; forecasting; volatility; emerging market; GARCH models; root mean square error; RMSE.
Afro-Asian Journal of Finance and Accounting, 2019 Vol.9 No.4, pp.363 - 380
Received: 04 Dec 2017
Accepted: 27 Jun 2018
Published online: 01 Oct 2019 *