Structured Monte Carlo

[Angshuman]

Hi David,
Thanks for your reply for the MCS. Requesting you to clarify the following doubts:

1. Another doubt for the ‘Structured Monte Carlo VaR ’ is that it uses correlation matrix which is derived from the historically observed returns. So even in Monte Carlo Simulation, historically observed events have critical influence. Hence events, which has never been observed in history (e.g recent happenings like 25 standard deviations!) can never be predicted either by Monte Carlo or by Historical Simulation. If this inference is right then what is the way out? How can we still use Monte Carlo to compute VaR?

2. Post your advice for the MCS, in the worksheet provided by you for ‘Structured Monte Carlo’ for portfolio VaR computation I used TINV(RAND(),6) instead of NORMSINV(RAND()) is this the correct way to get rid of normality assumption? But when I used TINV(RAND(),6), most of the random numbers are in positive zone so the resultant VaR is not correct. Can you please suggest the correct one?

 

[David Harper CFA FRM]

Hi Angshuman,

1. Right, but this "structured MCS" is merely the one basic flavor of Monte Carlo that we study in the FRM. I would recommend to you Carol Alexander's Book IV, she devotes space to MCS: http://www.amazon.com/Market-Risk-Analysis-Value-Models/dp/0470997885/ref=sr_1_2?ie=UTF8&s=books&qid=1249608441&sr=8-2
Or, I think Wilmott's book probably introduces. Just as we use a trivial normal inverse transform (=NORMSINV(RAND())), there is virtually no limit on the "engine" (the specification of the stochastic process). You can, e.g., model abrupt regime shifts. Giant topic, but not in FRM. Regarding correlation matrix, please note even with the our "simple" usage (covariance matrix), a typical "stress test" would stress the matrix.

2. I answered on that thread...

David