Usage: |
VaRMC = VaRestMC(VaRdelta,VaRgamma,VaRcovmatrix, smethod, opt)
|
Input: |
| VaRdelta | m x 1 vector of first derivatives, aggregated delta matrix
|
| VaRgamma | m x m Hessian matrix, aggregated gamma matrix
|
| VaRcovmatrix | m x m covariance matrix, variance-covariance matrix for the underlying portfolio
|
| smethod | Monte-Carlo sampling method for VaR. Default set at "IS".
"PS": Plain vanilla sampling method
"MS": Moment matching sampling method
"SS": Stratified Latin Hypercube sampling method
"IS": Importance sampling method
|
| opt | optional, a list with optional input. The function
"VaRopt" can be used to set up this parameter.
The order of the list elements is not important.
|
| opt.alpha | Significance level for VaR estimation. (Default = 0.01)
|
| opt.days | Estimation time horizon. (Default = 1 day)
|
| opt.nsimu | Number of Monte-Carlo simulations. (Default = 1000)
|
Output: |
| VaRMC | Estimated VaR |