Download as pdf or txt
Download as pdf or txt
You are on page 1of 1

Decomposition of VaR in Historical Simulation

In the parametric linear VaR framework there are analytic formulae that can be applied to obtain the gradient vector. But in the historical VaR model there are no such analytical formulae. So we estimate the gradient vector using a first order finite difference. We make a small perturbation in each of the risk factor, in turn, and compute the first partial derivative of the total VaR with respect to that risk factor by dividing the resulting change in the total VaR by the small perturbation. For each risk factor or component, we need to repeat the following steps: 1. Make a small perturbation So, if and in the returns of the risk factor or component are the returns and the weights of the

components of the portfolio, respectively. Then, for ith component, we need to calculate the shift in the component as given below:

2. Calculate the portfolio return with this shift

3. Revaluate the portfolio VaR with this shift

4. Find the difference between the revaluated portfolio VaR and the original portfolio VaR

5. To obtain the marginal VaR, divide the difference by the small perturbation

You might also like