風險值為應用於評估市場風險的風險管理技術，一個投資組合的風險值為其隨著持有期間變化，給定既定的百分比，對此投資組合價值變化的機率分配做簡單的估計。本文的研究使用變異數-共變異數法來計算風險值，並加入區分多空市場之概念，以GARCH-Normal、GARCH-t 估計風險值，模型中以馬可夫轉換模型、日報酬、月報酬、以及技術分析作為區分多空市場之依據。本研究以台灣加權股價指數作為研究對象，由失敗率、Kupiec（1995）、Christoffersen（1998）涵蓋比率LR 檢定結果顯示，相較於傳統計算風險值的模型，在模型中加入區分多空頭的虛擬變數，確實能夠有效的降低模型預測失敗次數。

Value at Risk（VaR）is a risk-management technique that has been widely used to assess market risk. VaR of a portfolio is an estimation of a specified percentile of the portfolio loss over a given holding period. In this study, we uses the variance-covariance method to calculate VaR. The purpose of this study is to assess the GARCH-Normal model and GARCH-t model for measuring VaR under bull and bear markets. The distinction of bull or bear market is according to four methods based on daily return,
monthly return, moving average, and Markov-Switching model, respectively. Based on failure rates, Kupiec（1995）and Christoffersen（1998）LR test, to separate market conditions to bull and bear markets could help to make the results more precise.