한국의 주요 환율 변동성과 글로벌 불확실성: 다변량 확률변동성 분석

초록

This study applies multivariate stochastic volatility (MSV) models with common factors to analyze the volatility of Korean exchange rate returns against the U.S. dollar, Japanese yen, euro, and Chinese yuan. I consider three specifications: a single-factor MSV model, a two-factor MSV model, and an extended single-factor MSV model that allows for idiosyncratic stochastic volatility at the country level. The models are estimated using a Bayesian Markov chain Monte Carlo (MCMC) method, and model adequacy is evaluated based on diagnostics of standardized residuals. The empirical results show that a substantial portion of the volatility of major exchange rates against the Korean won is explained by a common factor associated with global and U.S. economic conditions. However, the common-factor structure alone is insufficient to fully capture the dynamic behavior of the variance–covariance matrix. Allowing for time-varying idiosyncratic volatility for each exchange rate improves the overall model fit more effectively than introducing additional latent common factors. The estimated common factor volatility is found to have statistically significant correlations with the VIX and the economic policy uncertainty (EPU) indices of both Korea and the U.S. These results indicate that exchange rate volatility in Korea is driven not only by domestic factors but also by global financial market risk and economic policy uncertainty, particularly those originating from the U.S.

키워드