如表3.2所示，历史模拟下的VaR违例次数为34，与正常情况下的次数相

As shown in Table 3.2, the number of violations VaR historical simulation 34, the same number of times under normal conditions, and also exceeded the acceptable level of 2.53 days. When the value of the portfolio plummeted in the fourth quarter and in January 2008, estimated by the two methods are continuously underestimated VaR loss. Therefore, regardless of which calculation method you choose, in extreme adverse circumstances, VaR is not the appropriate method to predict potential losses.

As shown in Table 3.2, the number of VaR violations in the historical simulation is 34, the same as normal, and exceeds the acceptable level of 2.53 days. When portfolio values plummeted in January and the fourth quarter of 2008, VaR, which is estimated in both ways, continuously underestimated losses. Therefore, regardless of which calculation method is chosen, VaR is not the appropriate way to predict potential losses in extremely adverse circumstances.

As shown in table 3.2, the number of VaR violations in the historical simulation is 34, the same as in the normal case, and it also exceeds the acceptable level of 2.53 days When portfolio values plummeted in January and the fourth quarter of 2008, VAR estimates using both methods consistently underestimated losses Therefore, no matter which calculation method is chosen, VaR is not a suitable method to predict the potential loss in the extremely unfavorable situation.<br>