Dependence between stock market and foreign exchange market in South Asia: A Copula-Garch approach

Abstract

The paper studies the dependence pattern between stock market and foreign exchange market of three South Asian countries; namely Bangladesh, India and Sri Lanka by using five copula functions, to reveal asymmetric dependence structure. This paper focuses South Asia because of its promise as portfolio investment destination. The dependence structure between stock market and foreign exchange market assists MNCs, private equity firms, international portfolio managers and policy makers in international investment decision making. The paper also studies tail dependence (upper and lower). Markets crash together rather boom together, which can be traced by tail dependence parameter. Correlation, Kendall’s tau and Spearman’s rho reveal linear dependence. However, Copulas reveal non-linear dependence between non-normal distributions. In this paper, Copula functions are applied between marginal distributions of stationary uni-variate return series. Using daily return series for the period of July 31, 2009 to July 31, 2013, ARMA-GARCH type model are applied to obtain marginal distributions of return series. The results from marginal models indicate that volatility dies immediately after a crisis; meanwhile positive news creates more volatility than negatives. The results from copula models indicate existence of asymmetric dependence, with upper tail dependence for all pairs, implying dependence increases in bull market situation (price increase in financial instruments). Both Bangladeshi and Indian pairs provide some diversification possibility, against no diversification for investing in Sri Lankan market. Moreover, Bangladesh and India demonstrated as promising investment destination in terms of risk –return criteria. Foreign investors are recommended to employ proper risk and investment management strategy in order to earn capital gain, by following diversification strategy. Conversely, investment recipient countries are recommended to establish a stable, well regulated stock market, with high stock market capitalization to GDP. Besides, monetary policy should stabilize the exchange rate of currency to attract foreign investors. Copula based dependence measures assist foreign investors and investment recipient countries to change their related strategies and policies; subsequently they will be able to maximize their utility functions.

Sri Lankan Stock Market Volatility Analysis: An Arma- Garch Approach

“…Contrastingly, innovations of the model are negative and negatively decreasing suggesting the decreasing impact from the innovations or system shocks on the future stock returns. The behaviour of returns and innovations estimated are complementary to the findings of Kamal and Haque (2016) on Sri Lankan stock market behaviour post crisis. Moreover, the rapid decline of the prior returns and innovation on future stock returns provide confirmation to the findings of Arsalan, et.…”

“…Further, Sans-Cipollitti (2020) identified that ARMA-GARCH approach comparatively provided higher accuracy in stock index prediction. Kamal and Haque (2016) modeled the volatility of the Indian, Bangladesh and Sri Lankan stock markets in ARMA-GARCH to determine the fact that in the South Asian markets' volatility subsided promptly post crisis. Locally, Zhang and Nadarajah (2018) analysed CSE returns from 1985 to 2017 to conclude that the ARMA-APARCH modelling is efficient in forecasting volatility of the Sri Lankan market based on its log likelihood.…”

Dependence Structure Between Renminbi Movements and Volatility of Foreign Exchange Rate Returns

“…This study employs the copula-TARCH method for considering such asymmetric dependence. As shown in past studies (e.g., Kamal and Haque, 2016; Kresta 2013), this class of methods has good statistical performance when the process follows a distribution with fat tails and displays volatility clustering, which is the case for currency returns. As the purpose of this article is to examine currency risks due to unexpected and excessive RMB movements, tail dependence remains the focus of analysis.…”

“…In recent years, a number of researchers have started to extend the GARCH method to the copula-GARCH framework, which has a better ability to capture the dependence structure between two financial assets, by taking into account their volatility. Kamal and Haque (2016) applied the copula-GARCH approach to study the asymmetric dependence structure between daily returns of benchmark stock index and exchange rate in India, Bangladesh and Sri Lanka. Their results showed that both stock and foreign exchange markets suffer higher volatility when market news are positive.…”

Heterogenous responses of stock markets to covid related news and sentiments: Evidence from the 1st year of pandemic

“…Exchange rate uncertainty: Exchange rates affects stock returns negatively through capital mobility 5 ( Frankel, 1983 ; Liang et al, 2013 ; Kamal and Haque, 2016 ). Exchange rate volatility affects stock return volatility ( Kasman et al, 2011 ).…”