Professional Documents
Culture Documents
07 Texas Klugman 28
07 Texas Klugman 28
07 Texas Klugman 28
Stuart Klugman ()
Credibility Theory
1 / 62
Table of Contents
1
A History of Credibility
Types of Credibility
Limited Fluctuation Credibility
Greatest Accuracy Credibility
Credibility Example
Credibility Theory
2 / 62
Stuart Klugman ()
Credibility Theory
3 / 62
Credibility Theory
4 / 62
Credibility Theory
5 / 62
Stuart Klugman ()
Credibility Theory
6 / 62
Denition
Credibility is a measure of the predictive value in a
given application that the actuary attaches to a
particular body of data.
Denition
Full credibility is the level at which the subject
experience is assigned full predictive value based on a
selected condence interval.
Stuart Klugman ()
Credibility Theory
7 / 62
ASOP
Stuart Klugman ()
Credibility Theory
8 / 62
ASOP
Stuart Klugman ()
Credibility Theory
9 / 62
ASOP
Credibility Theory
10 / 62
A History of Credibility
A History of Credibility
Stuart Klugman ()
Credibility Theory
11 / 62
A History of Credibility
Introduction
Credibility Theory
12 / 62
A History of Credibility
Introduction
Stuart Klugman ()
Credibility Theory
13 / 62
A History of Credibility
Credibility Theory
14 / 62
A History of Credibility
Stuart Klugman ()
Credibility Theory
15 / 62
A History of Credibility
Stuart Klugman ()
Credibility Theory
16 / 62
A History of Credibility
Modern results
Stuart Klugman ()
Credibility Theory
17 / 62
Types of Credibility
Types of Credibility
Stuart Klugman ()
Credibility Theory
18 / 62
Types of Credibility
A List of Denitions/Types
Stuart Klugman ()
Credibility Theory
19 / 62
Types of Credibility
Types of Credibility
Limited Fluctuation
Stuart Klugman ()
Credibility Theory
20 / 62
Types of Credibility
Full Credibility
Denition
Full Credibility is assigned to a data set when the
probability that the relative error in the estimate is less
than r is at least p.
This is a condence interval approach in that it assigns
full credibility when a 100p% condence interval for the
relative error has a width that is less than 2r. For
example, we may want to be 90% condent that the
relative error is less than 5%. Then p = 0.9 and
r = 0.05.
Note - the exposure method (dollars, lives, etc) is not
relevant. What you use for an estimator is relevant.
Stuart Klugman ()
Credibility Theory
21 / 62
Types of Credibility
An Example
Example
We want to estimate the mortality rate, q for a group of
lives ages 50-59. We have one year experience on each of
1,000 lives. For each life,j, we have recorded the amount
of insurance, bj . We have also recorded the outcome,
dj = 1 if the life died, dj = 0 if the life lived. We choose
1000
the dollar weighted estimate, q = 1000
j =1 bj dj j =1 bj . Is
this fully credible with p = 0.9 and r = 0.05?
We need to calculate Pr(j(q
it is greater than 0.9.
Stuart Klugman ()
Credibility Theory
22 / 62
Types of Credibility
Three assumptions
2
3
Stuart Klugman ()
Credibility Theory
23 / 62
Types of Credibility
Example, continued
Stuart Klugman ()
Credibility Theory
q ).
24 / 62
Types of Credibility
Example, continued
q
q
< 0.05
q < 0.05q )
0.05q
0.05q
<Z <
Stuart Klugman ()
Credibility Theory
25 / 62
Types of Credibility
Example, concluded
Credibility Theory
26 / 62
Types of Credibility
Partial credibility
Stuart Klugman ()
Credibility Theory
27 / 62
Types of Credibility
Example
Example
Determine the weight for the previous example
The goal is to get the right hand term in the probability
statement to be 1.645. The equation to solve is:
0.05q
1.645 =
=h
Stuart Klugman ()
0.05q nj=1 bj
q (1
Credibility Theory
q ) nj=1 bj2
i1/2 .
28 / 62
Types of Credibility
Example, continued
1.645 =
Stuart Klugman ()
Credibility Theory
29 / 62
Types of Credibility
Example, continued
Stuart Klugman ()
Credibility Theory
30 / 62
Types of Credibility
PROS:
Good for experience rating, where there is a default
premium.
Simple to implement and understand.
CONS:
Reects only reliability of data, not of base rate.
May not have an obvious base rate.
No sound statistical justication.
Stuart Klugman ()
Credibility Theory
31 / 62
Types of Credibility
Types of Credibility
Greatest Accuracy
Stuart Klugman ()
Credibility Theory
32 / 62
Types of Credibility
Statistical model
Stuart Klugman ()
Credibility Theory
33 / 62
Types of Credibility
Myths
Stuart Klugman ()
Credibility Theory
34 / 62
Types of Credibility
Bhlmann credibility
Z =
Credibility Theory
35 / 62
Types of Credibility
Bhlmann credibility
Stuart Klugman ()
Credibility Theory
36 / 62
Types of Credibility
Credibility Example
Types of Credibility
An Example
Stuart Klugman ()
Credibility Theory
37 / 62
Types of Credibility
Credibility Example
A baseball example
Example
As of May 30, 2006, 77 National League batters had 175
or more plate appearances. Their batting averages
ranged from Miguel Cabrera (.346) to Clint Barmes
(.191). Estimate the 77 season-ending averages and
compare the answers to the end-of-season actual
averages.
Stuart Klugman ()
Credibility Theory
38 / 62
Types of Credibility
Credibility Example
Stuart Klugman ()
Credibility Theory
39 / 62
Types of Credibility
Credibility Example
Stuart Klugman ()
Credibility Theory
40 / 62
Types of Credibility
Credibility Example
Stuart Klugman ()
Credibility Theory
41 / 62
Types of Credibility
Credibility Example
Conclusions
Stuart Klugman ()
Credibility Theory
42 / 62
Stuart Klugman ()
Credibility Theory
43 / 62
The problem
Stuart Klugman ()
Credibility Theory
44 / 62
The estimator
Let
n
g
g
g
g
i =1 j =1 bij fij dij
i =1 j =1 bij fij dij
m
= g
=
ng
e
i =1 j =1 bij fij qis
Stuart Klugman ()
Credibility Theory
45 / 62
Stuart Klugman ()
Credibility Theory
46 / 62
g
i =1
!
ng
b
f
d
j =1 ij ij ij
e
ng
i =1 j =1 bij fij qi
=
e
2 = Var (m
) = Var
g
Stuart Klugman ()
i =1 qi (1
g
i =1
!
ng
b
f
d
j =1 ij ij ij
e
qi ) (bij fij )2
.
e2
Credibility Theory
47 / 62
m
gi=1 qis j =g 1 bij fij
=
=m
e
g
is (1 mq
is ) (bij fij )2
i =1 mq
2
=
e2
Alternatives would be to use the sample qi values are the
standard table qis values without adjustment.
Stuart Klugman ()
Credibility Theory
48 / 62
Obtaining Z
rm
Stuart Klugman ()
Credibility Theory
Z)
49 / 62
An Example
Stuart Klugman ()
Credibility Theory
50 / 62
An Example
The data
Stuart Klugman ()
Credibility Theory
51 / 62
An Example
Stuart Klugman ()
0.7306 = 0.9638.
Credibility Theory
52 / 62
Conclusions
Stuart Klugman ()
Credibility Theory
53 / 62
Conclusions
The Good
Stuart Klugman ()
Credibility Theory
54 / 62
Conclusions
The Bad
Stuart Klugman ()
Credibility Theory
55 / 62
Conclusions
The Ugly
Stuart Klugman ()
Credibility Theory
56 / 62
Conclusions
Stuart Klugman ()
Credibility Theory
57 / 62
A Non-credibility Approach
Stuart Klugman ()
Credibility Theory
58 / 62
A Non-credibility Approach
Problem
What mortality ratio should we use when data are
limited?
How about a two-step process?
1
2
If m
1.96 includes 1, use the standard table.
if above,
If the interval is below 1, use m
+ 1.96,
use m
1.96
Credibility Theory
59 / 62
A Non-credibility Approach
Stuart Klugman ()
Credibility Theory
60 / 62
A Non-credibility Approach
Example re-visited
Stuart Klugman ()
Credibility Theory
61 / 62
A Non-credibility Approach
Stuart Klugman ()
Credibility Theory
62 / 62