如何用matlab求风险价值,Var风险价值的计算方法

1. Find a set of return data from

a set of price/rate data: = P2/P1 - 1 or Ln(P2) - Ln(P1)

2. Calculate mean = AVERAGE(return data)

3. Calculate stdev = STDEV(return data)

4. Calculate VaR (99% confidence level) = mean - 2.33*stdev, should

be negative.

5. today value = latest price/rate data

6. tomorrow worst case = today value* (1 + VaR)

note: daily data of one asset

If a portfolio of assets,

The steps:

1. Collect historical daily data for asset A prices and calculate

the daily return.

2. Collect historical daily date for asset B prices and calculate

the daily return.

3. Generate covariance matrix for asset A and B.

4. Calculate portfolio variance and standard deviation. In Excel,

it can be done using MMULT(MMULT(TRANSPOSE(value),covariance)

5. Calculate VaR at 99% confidence level as 2.326 multiply by

standard deviation.

6. Tomorrow's worst value = Today's value - VaR

以上是如何在excel中处理var模型的计算。要点在于计算方法的学习。同样，在stata中也可以作同样的处理操作，然而对于矩阵的计算，matlab更为强大一些。