Regime-Switching volatility and risk quantification in South Asian and developed stock Markets: A Comparative perspective using Markov-Switching GARCH with MLE and MCMC estimations

This study investigates volatility regime dynamics and risk quantification across the developed stock markets of the NYSE and SSEC, and the emerging markets of South Asia, using the Markov-Switching GARCH framework. By employing both Maximum-Likelihood Estimation (MLE) and Bayesian Markov Chain Monte Carlo (MCMC) methods, the study captured volatility clustering that depends on regimes, their persistence, and transition probabilities. The findings of the MLE have revealed significant regime shifts in the markets of South Asia and have displayed frequent transitions, high volatility clustering, especially during low-volatility regimes, and a higher level of instability than in developed equity markets. Moreover, the MCMC findings further substantiate these findings by providing robust parameter estimates and revealing stronger volatility persistence during the calm regime and greater volatility persistence during turbulent periods in the developing South Asian stock markets. Then, volatility forecasting shows sustained market uncertainty, with emerging South Asian stock markets exhibiting higher volatility than developed markets. Moreover, the findings on Value-at-Risk (VaR) and Expected Shortfall (ES) have confirmed the elevated tail risk in the developing South Asian market, especially in Nepal and the Dhaka Stock Exchange. These findings contribute to the literature by providing an empirical comparison of risk and volatility across developed and developing markets, validating the efficiency of regime-switching models when combined with Bayesian estimation techniques for capturing the complex behaviour of financial markets.