本文採用RiskMetrics模型與GARCH模型及馬可夫轉換模型估算避險基金指數之風險值，並進一步以RiskMetrics模型與GARCH模型及馬可夫轉換模型所估出之風險值進行比較，用以探討何種模型有較佳的預測能力及績效，使投資大眾於面臨風險時，能正確的評估與控管，以避免承擔超過預期的損失，實證結果如下： 1.由回溯測試的結果可知，RiskMetrics模型與GARCH模型及馬可夫轉換模型都能有效的估計風險值，風險控管能力均有一定的水準，其中又以馬可夫轉換模型在信心水準99%表現最佳。 2.就均方差標準根檢定而言，馬可夫轉換模型的表現為三模型中最優異的，推斷其原因為馬可夫模型採用馬可夫鏈做為狀態轉換的機制，相較於RiskMetrics模型與GARCH模型，更能夠考慮資料序列前後期狀態與相關訊息，進而對報酬分配有較精確的掌握。 This paper investigates the Valu-at-Risk(VaR) of returns on hedge fund indexes using the RiskMetrics , the GARCH and the Markov Switching Models. Furthermore, we compared the Valu-at-Risk(VaR) between the RiskMetrics , the GARCH and the Markov Switching Models. The purpose is to find out which of three models has better prediction and performance for investors to evaluate and to take control in order to avoid unexpected lost while minimizing damage. The result of this study shows the following: 1.The back-test shows that the RiskMetrics model, the GARCH model and the Markov Switching Model can estimate Valu-at-Risk(VaR) effectively which proves that the ability to control risk is at a good standard. Besides, the empirical results show Markov Switching Model can capture the distribution better than the others with a 99% confidence level under the back-test. 2.According to the RMSE, the Markov Switching Model erforms better than either the GARCH model or the RiskMetrics model. We infer that the Markov Switching Model can well capture the distribution resulting from the adoption of the transformation mechanism of Markov chain. The Markov chain contains more relative information of time serial data than other models do.
