VOLATILITY MEASUREMENT IN INDIA FOREX MARKET USING GARCH MODAL

Abstract

This dissertation is a study on volatility measurement of Foreign exchange market in India using GARCH model. This study includes an overview of Indian Foreign exchange market and its position with respect to global Forex market .Regimes of Indian foreign exchange market have been studied to bring out the impact of high volatility on the foreign trade and economic growth in India .The periods of high volatility have causes a substantial decrease in foreign trade and economic activity in the country emphasizing the importance to forecast volatility so that the competent authority can take corrective measures to check high volatility. Different models that had been used to forecast volatility by researchers have been summarized in the literature review. Also in the literature review have been analyzed that GARCH model gives more accurate results in forecasting volatility than the other available models. The objectives of this study are to (a) Summarize different models available for forecasting volatility (b) Forecast volatility of Indian foreign exchange market using GARCH model Vast literature on the subject of volatility measurement of Forex have justified that volatility can be expressed as conditional variance and time series data modeling can be used to measure volatility. Some models use standard deviation to predict volatility that gives biased results. GARCH model use conditional variance and many researchers have studied the accuracy of volatility measurement using GARCH model in other foreign countries and found that it gives satisfactory results with the use of constraints of stationarity. A number of different types of GARCH models have been developed for improving the accuracy of volatility forecast. INR and USD currency pair data from January 2007-January 2017 is used for this study as it is the currency pair in which major part of foreign exchange trading in India is done. Analysis of volatility forecast by GARCH model shows that although the errors are not normally distributed but the estimators of volatility are consistent and GARCH model can be satisfactorily used for volatility forecast of Indian foreign exchange market