Now from the above definitions and assumptions we have Cov bracketleftbig μ Θ j

Now from the above definitions and assumptions we

This preview shows page 3 - 6 out of 15 pages.

. Now from the above definitions and assumptions we have Cov bracketleftbig μ j ) , X i prime r prime bracketrightbig = E braceleftbig Cov bracketleftbig μ j ) , X i prime r prime | Θ j bracketrightbigbracerightbig +Cov braceleftbig E [ μ j ) | Θ j ] , E ( X i prime r prime | Θ j ) bracerightbig , = 0 + Cov braceleftbig μ j ) , E ( X i prime r prime | Θ j ) bracerightbig , (4.4) where E ( X i prime r prime | Θ j ) = braceleftbigg E ( X i prime r prime ) = m , if i prime negationslash = j μ j ) , if i prime = j . Hence, substituting back in (4.4) we get Cov bracketleftbig μ j ) , X i prime r prime bracketrightbig = braceleftbigg Cov braceleftbig μ j ) , m bracerightbig = 0 , if i prime negationslash = j Cov braceleftbig μ j ) , μ j ) bracerightbig = V bracketleftbig μ j ) bracketrightbig = a , if i prime = j = δ i prime j a , where δ ij = braceleftbigg 1 if i = j 0 if i negationslash = j ,
Image of page 3

Subscribe to view the full document.

38 CHAPTER 4. B ¨ UHLMANN’S CREDIBILITY MODELS is Kronecker’s symbol. Similarly Cov( X ir , X i prime r prime ) = E braceleftbig Cov( X ir , X i prime r prime | Θ i ) bracerightbig + Cov braceleftbig E ( X ir | Θ i ) , E ( X i prime r prime | Θ i ) bracerightbig , = δ ii prime ( δ rr prime s 2 + a ) , which allows us to re–write (4.3) as δ i prime j a = k summationdisplay i =1 n summationdisplay r =1 c j ir δ ii prime ( a + δ rr prime s 2 ) , = n summationdisplay r =1 c j i prime r ( a + δ rr prime s 2 ) , (4.5) = a c j i prime . + c j i prime r prime s 2 , (4.6) for every i prime = 1 , . . . , k and r prime = 1 , . . . , n, where c j i prime . = n r =1 c j i prime r . Summing on both sides with respect to r prime we obtain δ i prime j a n = a c j i prime . n + s 2 c j i prime . , for every i prime = 1 , . . . , k, that is c j i prime . = δ i prime j a n s 2 + a n = δ i prime j Z , where Z = a n s 2 + a n . A substitution of this final expression in (4.6) gives c j i prime r prime = δ i prime j (1 - Z ) a s 2 , for every i prime = 1 , . . . , k and r prime = 1 , . . . , n . We see that all these coefficients are 0 unless i prime = j and that the optimal estimator is thus c j 0 + k summationdisplay i =1 n summationdisplay r =1 c j ir X ir = s 2 s 2 + a n m + n summationdisplay r =1 [(1 - Z ) a s 2 X jr ] = (1 - Z ) m + Z ¯ X j a50 Although called a credibility “estimator”, the above premium formula depends on the portfolio parameters m , s 2 and a , which have yet to be estimated by some distribution free approach.
Image of page 4
4.1. THE CLASSICAL MODEL OF B ¨ UHLMANN (1967) 39 Lemma 4.1. The following estimators ˆ m = ¯ X = 1 k k summationdisplay j =1 ¯ X j = 1 kn k summationdisplay j =1 n summationdisplay r =1 X jr , ˆ s 2 = 1 k ( n - 1) k summationdisplay j =1 n summationdisplay r =1 ( X jr - ¯ X j ) 2 , and ˆ a = 1 ( k - 1) k summationdisplay j =1 ( ¯ X j - ¯ X ) 2 - 1 n ˆ s 2 , are unbiased for m , s 2 and a , respectively. Proof: [De Vylder (1981)] The first assertion E ( ˆ m ) = m is trivial. Then, from classical statistics for iid observations it is clear that for every fixed j = 1 , . . . , k . 1 ( n - 1) n summationdisplay r =1 ( X jr - ¯ X j ) 2 is a conditionally unbiased estimator of σ 2 j ) = V ( X jr | Θ j ), which means that 0 = E bracketleftBig 1 ( n - 1) n summationdisplay r =1 ( X jr - ¯ X j ) 2 - σ 2 j ) | Θ j bracketrightBig .
Image of page 5

Subscribe to view the full document.

Image of page 6
  • Fall '09
  • Dr.D.Dryanov
  • Trigraph, Estimation theory, Mean squared error, Bias of an estimator, Credibility Models

What students are saying

  • Left Quote Icon

    As a current student on this bumpy collegiate pathway, I stumbled upon Course Hero, where I can find study resources for nearly all my courses, get online help from tutors 24/7, and even share my old projects, papers, and lecture notes with other students.

    Student Picture

    Kiran Temple University Fox School of Business ‘17, Course Hero Intern

  • Left Quote Icon

    I cannot even describe how much Course Hero helped me this summer. It’s truly become something I can always rely on and help me. In the end, I was not only able to survive summer classes, but I was able to thrive thanks to Course Hero.

    Student Picture

    Dana University of Pennsylvania ‘17, Course Hero Intern

  • Left Quote Icon

    The ability to access any university’s resources through Course Hero proved invaluable in my case. I was behind on Tulane coursework and actually used UCLA’s materials to help me move forward and get everything together on time.

    Student Picture

    Jill Tulane University ‘16, Course Hero Intern

Ask Expert Tutors You can ask 0 bonus questions You can ask 0 questions (0 expire soon) You can ask 0 questions (will expire )
Answers in as fast as 15 minutes