This preview shows pages 1–3. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Review Notes for Loss Models 1  ACTSC 431/831, FALL 2008 Part 4 Frequency and Severity with Coverage Modifications Let X be the groundup loss for an insurance policy or an insurer and assume that X is a continuous r.v. with cdf F ( x ), sf S ( x ), and pdf f ( x ). 1. An ordinary deductible of d : With the deductible, the policy pays nothing if X d and pays X d if X > d . (a) The perpayment random variable Y P is given by Y P = undefined , X d X d, X > d = X d  X > d. (b) The perloss random variable Y L is given by Y L = , X d X d, X > d = ( X d ) + = X X d. (c) The mean of Y P is given by E ( Y P ) = E ( X d  X > d ) = e ( d ) = R d S ( x ) dx 1 F ( d ) . (d) The mean of Y L is given by E ( Y L ) = E [( X d ) + ] = E ( X ) E ( X d ) = Z d S ( x ) dx. (e) The relationship between the moments of Y P and Y L : E Y P k = E Y L k 1 F ( d ) , k = 1 , 2 ,.... (f) The perpayment random variable Y P is also called excess loss variable or left truncated and shifted variable and the perloss random variable Y L is also called left censored and shifted variable. They have been discussed in detail in Part 1. (g) E ( Y L ) and E ( Y P ) are also called expected cost per loss and expected cost per payment , respectively, for the ordinary deductible. (h) See Part 1 for distributions and more results about Y P and Y L . 1 2. A franchise deductible of d : With the deductible, the policy pays nothing if X d and pays X if X > d . (a) The perpayment random variable Y P f is given by Y P f = undefined , X d X, X > d = X  X > d. (b) The perloss random variable Y L f is given by Y L f = , X d, X, X > d. (c) The perpayment random variables Y P and Y P f are conditional positive random variables and defined only for X > d . In particular, Y P > 0 and Y P f > d hold. (d) The perloss random variables Y L and Y L f are mixed nonnegative ran dom variables and defined for all X . In particular, Pr { Y L = 0 } = Pr { Y L f = } = Pr { X d } = F ( d )....
View Full
Document
 Fall '09
 david

Click to edit the document details