

FlashChat  Actuarial Discussion  Preliminary Exams  CAS/SOA Exams  Cyberchat  Around the World  Suggestions 


Thread Tools  Search this Thread  Display Modes 
#205




Page 11 of Brosius, I'm looking at the linear approximation formula and the 1.2.3 listed for Hugh White's questions. If Cov(X,Y) < Var(X), isn't L(x) still bigger than E[Y]? How is this a decrease in the reserve? Isn't this true for all conditions?
Like, let's say x = 5 and E(X) = 4, then under all conditions E[YX] = (something bigger than 0 ) + E[Y]
__________________
Last edited by AbedNadir; 12112019 at 10:36 AM.. 
#207




Mack 1994 is bothering me.
on pg.112, Mack states The fact that the chain ladder estimator uses weights which are proportional to ��jk therefore means that ��jk is assumed to be inversely proportional to ������(Cj,k+1/CjkCj1,...,Cjk) I don't get how he came up with this conclusion. Why does the weight in the CL estimator imply Cjk is inversely proportional to the variance of the future development factor? 
#208




Quote:
So if you use a formula with a certain set of weights, you are implicitly making a statement about what you believe the variance relationship is (because if you believed it was something else, you are stupid for not using the MVUE for what you thought it was). 
#209




Quote:
Assume Cov(X,Y)/Var(X)=0.9, E[Y]=10 Expected reserve = E[Y]  E[X] = 10  4 = 6 Estimated Ultimate L(x) = (54)*0.9+10 =10.9 Reserve: L(x)  x = 10.95=5.9 < 6, hence the decrease 
#210




Quote:
__________________

Thread Tools  Search this Thread 
Display Modes  

