- Tedypendah
**Posts:**147**Joined:**

Hi all,I am busy with an exercise as part of building an Economic Capital model at an Insurance Company. Suppose I quantify my Lapse risk to be 2000, can someone confirm if I am right in assuming a normal distribution for the Lapse risk here. I think I read that one of the assumptions underlying the Solvency II standard formula is normality for individual risks but I can't seem to locate where I read it. 1) Is it enough to assume normality of individual risks when they are aggregated using a correlation matrix which means that a correlation matrix implies Pearson's rho? 2) Can we have a correlation matrix made up of Kendall's Tau?

Last edited by Tedypendah on March 29th, 2015, 10:00 pm, edited 1 time in total.

- Tedypendah
**Posts:**147**Joined:**

Quants please help, I know you know the answer. I'm an Actuary who is clueless here, see questions 1 and 2 Daveangel?

Last edited by Tedypendah on March 29th, 2015, 10:00 pm, edited 1 time in total.

- Tedypendah
**Posts:**147**Joined:**

hEEEEeeeeeelllllp, iyoooooooooooooooh, Heeeeeeelllllp, iyooooooh!

correlation is a linear measure of dependence,implicit assumption is normality.

- Tedypendah
**Posts:**147**Joined:**

Thanks Samsaveel, but Kendall's Tau is also correlation (rank)...im really confused. Or did you mean to say correlation matrix implies linear correlation i.e we can't use Kendall's Tau in a correlation matrix?

linear correlation is very confusing !it is used everywhere in finance , it only makes sense if the underlying RV ( Bivariate , Multivariate)are normal,or more generallycome from the elliptical world ,send me a pm ,will dig up a good paper that goes into details.

- Tedypendah
**Posts:**147**Joined:**

Thanks a lot Sam, will pm you.

Last edited by Tedypendah on April 2nd, 2015, 10:00 pm, edited 1 time in total.

GZIP: On