VAR

CFA Level III Derivatives
#1

It may be silly but I am freezing on the calculation.

In the CAFI 2013 Am exam, there is a question regarding VAR calculation.

Capital 30 sends its largest client a weekly VAR estimate using a probability of 5%. Currently this estimate is GBP 0.8 million, so Capital 30 has advised the client to be prepared for losses greater than this amount up to five weeks every three years.

The correction for the comment is :

The client should expect losses greater than GBP 0.8 million as often as 8 weeks every three years [calculated as 0.05 x 3 years x 52 weeks per year = 7.8].

My question is why did we add the 0.05 in the calculation. When we want to get the weekly var from the annual one we divide the E® by 52 and stdev. by (sq rt of 52) without including the 0.05.

#2

You’re thinking too much into this question, no mean-variance calculations are needed.

There are 52 * 3 =156 weeks in 3 years. Of those, the 5% VaR is the min amount you can expect to exceed in 7.8 weeks (8 rounded).

#3

I think converting the weekly VAR to annual would be incorrect, since the measurement interval should stay at 1 week and we would have ignored the confidence interval. Without changing the original measurement interval, we can think of 5% much like we do in historical VAR estimates, where we multiply the number of observations with the tail (5% here).