The International Portfolio Diversification under Extreme Markets

[[abstract]]本文主要探討國際投資組合處於極端市場時之關聯性及其風險分散效益的變化。本研究利用投資組合的風險值(value at risk, VaR)以及分位數相關係數，估算出特定信心水準下的條件相關係數，以衡量亞洲四小龍、新興金磚四國與歐美已開發國家等國際投資組合之間的極端報酬相關。實證發現雖然極端市場下之國際投資組合之間的相關性偏高，但是新興市場的投資組合與已開發國家的投資組合配對的相關性還是比較低的，尤其以納入金磚四國的投資組合相關性最低，顯示在極端市場時，投資人仍可使用亞洲新興股票市場及金磚四國投資組合以規避下方風險。此外，極端信賴水準下相關係數的信賴區間較寬，隱含此時的條件相關不容易準確地預測，所以調整投資組合時必須參酌更多資訊。本文的投資組合建構方式貼近組合基金的概念，且直接以投資組合報酬分位數估計條件相關，可以避免size-dependence 的偏誤，故研究結果可以提供投資者或基金經理人作為調整投資組合與風險控管的參考。[[abstract]]The main purpose of this study is to investigate the correlations across international portfolios and their diversification benefits in extreme market conditions This study measures the extreme return correlations between international portfolios including Asian emerging market portfolio BRICs portfolio and developed market portfolio using portfolio value at risk (VaR) and plots the 95 percent confidence intervals for the conditional correlations Empirical evidence shows while conditional correlations across international portfolios are higher under extreme market conditions the conditional correlation between emerging stock market portfolios and developed market portfolio is low especially the portfolio containing BRICs produces the lowest correlation This result implies that investors could avoid the downside risk and thus improve the benefit of portfolio diversification through portfolios constructed out of Asian emerging stock markets and BRICs stock index However the confidence interval of return correlation is getting wide at extreme confidence level implies that conditional correlation is hard to predict accurately In this study we simply use quantile of the portfolio returns to estimate conditional correlation This approach avoids size-dependent bias so that the empirical result is valuable to investors and money manager