01.08.2014 Views

Estimation, Evaluation, and Selection of Actuarial Models

Estimation, Evaluation, and Selection of Actuarial Models

Estimation, Evaluation, and Selection of Actuarial Models

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

14 CHAPTER 2. MODEL ESTIMATION<br />

i d i x i u i i d i x i u i<br />

1 0 - 0.1 16 0 4.8 -<br />

2 0 - 0.5 17 0 - 4.8<br />

3 0 - 0.8 18 0 - 4.8<br />

4 0 0.8 - 19—30 0 - 5.0<br />

5 0 - 1.8 31 0.3 - 5.0<br />

6 0 - 1.8 32 0.7 - 5.0<br />

7 0 - 2.1 33 1.0 4.1 -<br />

8 0 - 2.5 34 1.8 3.1 -<br />

9 0 - 2.8 35 2.1 - 3.9<br />

10 0 2.9 - 36 2.9 - 5.0<br />

11 0 2.9 - 37 2.9 - 4.8<br />

12 0 - 3.9 38 3.2 4.0 -<br />

13 0 4.0 - 39 3.4 - 5.0<br />

14 0 - 4.0 40 3.9 - 5.0<br />

15 0 - 4.1<br />

j y j s j r j<br />

1 0.8 1 32 − 0 − 2=30or 0+32− 0 − 2=30<br />

2 2.9 2 35 − 1 − 8=26or 30 + 3 − 1 − 6=26<br />

3 3.1 1 37 − 3 − 8=26or 26 + 2 − 2 − 0=26<br />

4 4.0 2 40 − 4 − 10 = 26 or 26 + 3 − 1 − 2=26<br />

5 4.1 1 40 − 6 − 11 = 23 or 26 + 0 − 2 − 1=23<br />

6 4.8 1 40 − 7 − 12 = 21 or 23 + 0 − 1 − 1=21<br />

¤<br />

Exercise 3 Repeat the above example, treating “surrender” as “death.” The easiest way to do this<br />

is to reverse the x <strong>and</strong> u labels <strong>and</strong> then use the above formula. In this case death produces censoring<br />

because those who die are lost to observation <strong>and</strong> thus their surrender time is never observed. Treat<br />

those who lasted the entire fiveyearsassurrendersatthattime.<br />

Despitealltheworkwehavedonetothispoint,wehaveyettoproduceanestimator<strong>of</strong>the<br />

survival function. The one most commonly used is called the Kaplan-Meier Product-Limit<br />

Estimator. Begin with S(0) = 1. Because no one died prior to y 1 , the survival function remains<br />

at 1 until this value. Thinking conditionally, just before y 1 , there were r 1 people available to die, <strong>of</strong><br />

which s 1 did so. Thus, the probability <strong>of</strong> surviving past y 1 is (r 1 − s 1 )/r 1 .Thisbecomesthevalue<br />

<strong>of</strong> S(y 1 ) <strong>and</strong> the survival function remains at that value until y 2 . Again, thinking conditionally,<br />

the new survival value at y 2 is S(y 1 )(r 2 − s 2 )/r 2 . The general formula is<br />

⎧<br />

1, 0 ≤ t

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!