AN ANALYSIS OF THE IMPACT OF SELECTED FOREIGN CURRENCY EXCHANGE RATES ON THE INDIAN STOCK MARKET USING SYMMETRIC AND ASYMMETRIC GARCH MODELS

Abstract

In the context of financial economics, the complex relationship between foreign currency exchange rates and stock prices are not easily comprehensible. The global integration of domestic and international financial markets has opened up new opportunities and consequently exposing the financial system to sizable risks in the Indian economy. The purpose of the study is to analyse the impact of different foreign exchange rates, viz USDINR, EURINR, GBPINR, JPYINR on the Nifty 50 index. This research is based on secondary data and quantitative in nature.  The data has been collected from the websites of Wall Street Journal, Bloomberg, investing.com and finance.yahoo.com. The weekly closing value of USDINR, EURINR, GBPINR, JPYINR and the Nifty-50 stock market index have been collected for a period of 16 years, from 24th September, 2007 to – 25th September, 2023. The four currency pairs, USDINR, EURINR, GBPINR, JPYINR have been taken as the sample to represent the FOREX market, and the index NIFTY-50 in India are examined. The econometric and statistical tools used to analyse the time series data in this research are (a) descriptive statistics (b) Unit Root test (stationarity) (c) cointegration test (d) regression analysis (e) Auto-correlation test, (f) heteroskedasticity test, (g) GARCH model (f) T-GARCH, (g) E-GARCH. The cointegration test reveals that there is no long-term association between NIFTY-50 and USDINR, EURINR, GBPINR and JPYINR. However, the Granger causality test finds short-term association between NIFTY-50 and USDINR, EURINR and JPYINR. The OLS regression analysis reveals that the beta of USDINR, EURINR, GBPINR and JPYINR are significant predictors of NIFTY-50. Moreover, the R-Square is found to be 30.47%, which shows the soundness of the regression fit. GARCH is a symmetric model and the analysis shows that the beta of GARCH (1,1) are significant. Therefore, it can be concluded that GARCH effects exist and it can be used to forecast the variance of NIFTY-50. The asymmetric GARCH models like TARCH and EGARCH reveals the presence of asymmetrical effect in the data series. Therefore, the investors have to be careful about the type of information with respect to USDINR, EURINR, GBPINR and JPYINR exchange rates in general and negative information in particular.