....

,...-

27. METHOD OF MOMENTS

456

Exercises
27.1.

Claim sizes are as follows:

100 200

500

1000

1500

2000

3700

An exponential is fitted to claim sizes using the method of moments.

Let X be the claim size.

Determine the estimate ofPr(X> 500).
27.2.

Claim sizes are as follows:

100

100

200

200

300

500

1000

2000

A two parameter Pareto distribution is fitted to the claim sizes using the method of moments. Let X be the claim
size.
Determine the estimate ofPr(X > 500).
27.3.

--

[C4 Saritple:8] Summary statistics for a sample of 100 losses are:

Interval

Number of Losses

(0, 2,000]
(2,000, 4,000]
(4,000, 8,000]
(8,000, 15,000]
(15,000, oo]

39
22
17
12

Total

Sum
38,065
-------63,816
96,447
137,595
331,831
667,754

10

100

Sum of Squares
52,170,078
194,241,387
572,753,313
1,628,670,023
17,906,839,238
20,354,674,039

A Pareto distribution is fit to this data using the method of moments.

Determine the parameter estimates.
27.4.

[4-800:36] You are given the following sample of five claims:

4

21

99

421

You fit a Pareto distribution using the method of moments.

Determine the 95th percentile of the fitted distribution.
(A)
(B)

(C)
(D)
(E)

Less than 380

At least 380, but less than 395
At least 395, but less than 410
At least 410, but less than 425
At least 425

27.5. You fit losses to a Pareto using the method of moments, matching on the first 2 moments. The sample
mean is 500.
For what values of the empirical variance l:Cxi- i) 2 /n will a fit to a Pareto be possible?

C/4 Study Manual-11th edition

Copyright 201 0 ASM

Exercises continue on the next page ...

(_

-~~

457

EXERCISES FOR LESSON 27

27.6.

[C-805:24] The following claim data were generated from a Pareto distribution:
130

20

350

218

1822

Using the method of moments to estimate the parameters of a Pareto distribution, calculate the limited
expected value at 500.
(A)
(B)
(C)
(D)
(E)
27. 7.

Less than 250

At least 250, but less than 280
At least 280, but less than 310
At least 310, but less than 340
At least 340
[4-F00:2] The following data have been collected for a large insured:

Year

Number Of
Claims

Average Claim
Size

1
2

100
200

10,000
12,500

'

Inflation increases the size of all claims by 10% per year.

A Pareto distribution with parameters a = 3 and 0 is used to model the claim size distribution.
Estimate 0 for Year 3 using the method of moments.
(B) 23,333

(A) 22,500

(C) 24,000

(D) 25,850

(E) 26,400

[4B-S98:5] (2 points) You are given the following:

27.8.

The random variable X has the density function

f(x)=a(x+ 1)-a-l, 0 <x < oo, a> 0.

A random sample of size n is taken of the random variable X.
Assuming a> 1, determine ii, the method of moments estimator of a.
(A)

27.9.

(B)

x-1

(C) X+I

(D) X-1

(E) X-!1

[4B-F95:17] (3 points) You are given the following:

Losses follow a Pareto distribution with parameters 0 (unknown) and a= 3.

300 losses have been observed.

Determine the variance of {j, the method of moments estimator of 0.

(A) 0.002502

(B) 0.003302

(C) 0.00500 2

(D) 0.01000 2

(E) 0.013302

(
C/4 Study Manual-lith edition
Copyright 2010 ASM

Exercises continue on the next page . ..

458

27. METHOD OF MOMENTS

27.10. Claim sizes are as follows:

10 20

30

40

50

A single-parameter Pareto distribution is fitted to the claim size using the method of moments. Both parameters,
a and (), are determined this way.
Determine the estimate of a.
27.11.

(4B-F97:18] (2 points) You are given the following:

The random variable X has the density function

f(x)= ax-a- 1 , 1 < x < oo,a > l.

A random sample is taken of the random variable X.

Determine the limit of ii, the method of moments estimator of a, as the sample mean goes to infinity.
(B) 1/2

(A) 0

(C) 1

(D) 2

(E)

oo

27.12. A study is performed of time from submission of a claim on a life insurance policy until payment. For
the 15 claims in the study, the time in days is as follows:
11
15

11
17

12
17

13
19

13
19

13
25

14
131

14

Using this data, the cumulative hazard rate at time 15 days, H(15), is estimated with 2 estimators:

(_

H1(15) is the Nelson-Aalen estimator.

H2 (15) is the method of moments estimator if the data are fit to a single parameter Pareto with 8 = 10.
Determine the absolute value of the difference between the estimates, IH1 (15)- H2 (15)j.
27.13. [Prior exam] A random sample of death records yields the following exact ages at death: 30, 50, 60, 60,
70, 90. The age at death of the population from which the sample is drawn follows a gamma distribution. The

parameters of the gamma distribution are estimated using the method of moments.
Determine the estimate of a.
(A) 6.0

27.14.
(i)
(ii)

(B) 7.2

(C) 9.0

(D) 10.8

(E) 12.2

[Prior exam] You are given:

Five lives are observed from time t = 0 until death.

Deaths occur at t = 3, 4, 4, 11, and 18.

Assume the lives are subject to the probability function

t e-t/c
f(t)=

-2-,

t > 0.

Determine c by the method of moments.

(A) !

(B) !

(C) 1

(D) 2

(E) 4

t
C/4 Study Manual-lith edition
Copyright 2010 ASM

Exercises continue on the next page ...

EXERCISES FOR LESSON 27

27.15.

459

Claim sizes are as follows:

10 20 40 80 100 200 200 500

A gamma distribution is fitted to the claim sizes using the method of moments.
Determine the estimate of().
27.16. [4B-S90:34] (2 points) The observations 1000, 850,750, 1100, 1250, and 900 are a random sample taken
from a gamma distribution with unknown parameters a and 8. Let a and {j denote the method of moments
estimators of a and e respectively,
1

In what range does

CAl
CBl
CCl
CDl
CEl
27.17.

a fall?

a< 30
30 ~a< 40
40~ a <50
so~a<60

60 ~a
[4B-S91:46] (2 points) The following is a sample of 10 claims:

1500
6000

3500
3800

1800
5500

4800
4200

3900
3000

The underlying distribution is assumed to be gamma, with parameters a and

e unknown.

In what range does the method of moments estimator, (}, of e fall?

(A)
(B)

(C)
(D)
(E)

27.18.

(} < 250
250 ~ (} < 300
300 ~ {) < 350
350 ~ {) < 400
400 ~ {)
[4B-S92:26] (1 point) The random variable X has the density function with parameter f1 given by
f(x; {1) = ; 2 xe-(xff1J2/Z,

x > 0,

f1 > 0,

whereE[X] = ~v'27[ and the variance ofXis2{1 2 - ~{1 2

You are given the following observations of X: 4.9, 1.8, 3.4, 6.9, 4.0.
Determine the method of moments estimate of fl.
(A)
(B)
(C)
(D)
(E)

\:

Less than 3.00

At least 3.00, but less than 3.15
At least 3.15, but less than 3.30
At least 3.30, but less than 3.45
At least 3.45

C/4 Study Manual-lith edition

Copyright 20 I 0 ASM

R'cercises continue on the next page ...

27. METHOD OF MOMENTS

460

[4B-S95:5] (2 points) You are given the following:

27.19.

The random variable X has the density function f(x) = axa-l, 0 < x < 1, a> 0.
A random sample of three observations of X yields the values 0.40, 0.70, 0.90.
Determine the value of a, the method of moments estimator of a.
(A)
(B)
(C)
(D)
(E)

Less than 0.5

At least 0.5, but less than 1.5
At least 1.5, but less than 2.5
At least 2.5, but less than 3.5
At least 3.5
[4B-F99:21] (2 points) You are given the following:

27.20.

The random variable X has the density function

f(x)= w fr(x)+(l- w)fz(x),

0<x

< oo, 0:::; w:::; 1.

A !lingle observation of the random variable X yields the value 1.

00

xfr(x)dx = l

J xf2(x)dx =2
00

fz(l)=2fi(l)r{;O

"'~-

Determine the method of moments estimate of w.

(B) l

(A) 0

27.2L

(C) l

(D) ~

(E) 1

[4B-S96:4] (2 points) You are given the following:

The random variable X has the density function f( X) =

:2 (e -X), 0 < X < e.

A random sample of two observations of X yields values 0.50 and 0.90.

Determine fJ, the method of moments estimator of e.
(A)
(B)
(C)
(D)
(E)
27.22.

Less than 0.45

At least 0.45, but less than 0.95
At least 0.95, but less than 1.45
At least 1.45, but less than 1.95
At least 1.95
A group of 3 individuals is hypothesized to have the following survival function:

scxJ= (w:xY

o:::;x:::;w

where w is a parameter.
Ages at death for the 3 individuals are 70, 60, and 80.
Estimate w using the method of moments.

L
C/4 Study Manual-!! th edition
Copyright 2010 ASM

Exercises continue on the next page ...

EXERCISES FOR LESSON 27

461

27.23. You observe the following ages at death:

60 70 75 80 86 87 88
You fit the survival function S(x) = ( 1- 1 ~0 )

8
,

0::; x::; 100 to this data using the method of moments.

Determine the fitted probability of survival past 80.

27.24.

On an insurance coverage, you observe the following claim sizes:

5

11

15

18

You fit an inverse exponential to this data using the method of moments, matching the first negative moment.
Determine the fitted probability of a claim higher than 10.
Losses have a mean of5 and a variance of15. You fit them to a mixture of2 uniform distributions, one of
them on [0, 01 ] with weight 0.6, and the other on [0, 02 ], 02 > 01 , with weight 0.4, using the method of moments.

27.25.

Determine 81.
27.26-27.

Use the following information for questions 27.26 and 27.27:

You are given the following:

The random variable X has the density function
f(x)=0.5 (

e-x/Bi) +0.5 ( i;-e-x!Bz ), 0 <X< 00, 0 < 01::; Oz.

A random sample taken of the random variable X has mean 1 and variance k.
27.26.

(A)
(B)
(C)

(D)
(E)

27.27.

[4B-F98:25] (3 points) If k is 3/2, determine the method of moments estimate of 01 .

Less than 1/5

At least 1/5, but less than 2/5
At least 2/5, but less than 3/5
At least 3/5, but less than 4/5
At least 4/5
[4B-F98:26] (2 points) Determine the values of k for which the method of moments estimates of 81 and

02 exist.
(A}

0<k

(B)

0< k <3
0< k <2

(C)

(D)

1::; k

(E)

1::; k <3

\.
C/ 4 Study Manual-11th edition
Copyright2010ASM

Exercises continue on the next page ...

27. METHOD OF MOMENTS

462

27.28.

[4-801:39] You are modeling a claim process as a mixture of two independent distributions A and B.

You are given:

(i) Distribution A is exponential with mean l.
(ii) Distribution B is exponential with mean 10.
(iii) Positive weight p is assigned to distribution A.
(iv) The standard deviation of the mixture is 2.

Determine p using the method of moments.

0.960

(A)

27.29.

(i)

(B)

0.968

0.972

(C)

(D)

0.979

(E)

0.983

[4-F01:33] You are given

Claim amounts follow a shifted exponential distribution with probability density function:
f(x)

1
= -e-<x-;t)fO

(}

A< X< oo

(ii) A random sample of claim amounts X1 ~X2 ~ .. ~X10 :

5 5 5 6 8 9 11
(iii)

12 16 23

l;Xi = 100 and l;X~ = 1306.

Estimate A using the method of moments.

(A) 3.0

(B) 3.5

(C)

4.0

(D) 4.5

(E)

5.0
-'-~

27.30.

[4-F03:24] You are given:

(i) A sample x 1

X 21

x 10 is drawn from a distribution with probability density function:

~ (~exp(-~) +~exp(-~)),
(ii)
(iii)

O<x<oo

()>a

l;xi = 150 and l;xf = 5000

Estimate () by matching the first two sample moments to the corresponding population quantities.
(A) 9

(B) 10

(C)

15

(D) 20

(E)

21

[4-F02:13] Losses come from an equally weighted mixture of an exponential distribution with mean m 1
and an exponential distribution with mean m 2

27.31.

Determine the least upper bound for the coefficient of variation of this distribution.
(A) 1

(B)

vz

(C)

;.13

(D) 2

(E)

.J5

(
C/4 Study Manual-11th edition
Copyright 2010 ASM

Exercises continue on the next page,,.

f~\

EXERCISES FOR LESSON 27

463

27.32.

[4-F03:8] For a sample of dental claims x 1 ,x2 , ,x10 , you are given:

L xi = 3860 and L xJ = 4,574,802

Claims are assumed to follow a lognormal distribution with parameters J-l and a.
(iii) J-l and a are estimated using the method of moments.
(i)

(ii)

Calculate E(X 1\ 500] for the fitted distribution.

(A)
(B)
(C)
(D)

(E)
27.33.

Less than 125

At least 125, but less than 175
At least 175, but less than 225
At least 225, but less than 275
At least 275
(4-F04:14) The parameters of the inverse Pareto distribution
F(x)= (xf(x+ e)f

are to be estimated using the method of moments based on the following data:
15

45

140

250

560

1340

Estimate 8 by matching kth moments with k = -1 and k = -2.

(A)
(B)

(C)
(D)

(E)
27.34.

Less than 1
At least 1, but less than 5
At least 5, but less than 25
At least 25, but less than 50
At least 50
[4-F04:24) You are given:

(i) Losses are uniformly distributed on (0, 8) with 8 > 150,

(ii) The policy limit is 150.
(iii) A sample of payments is:
14,33, 72,94, 120,135,150,150
Estimate e by matching the average sample payment to the expected payment per loss.
(A) 192

(B)

(C) 200

196

(D) 204

(E) 208

Additional released exam questions: C-F05:21, CAS3-F06:3

Solutions
27.1. For an exponential, set {j equal to the sample mean.
100+200+500+1000+1500+2000+3700 9000
=7
7
Pr(X > 500)= e-C.tioo)(soo) =I 0.67781
-

8=X=

d
'

C/ 4 Study Manual-11th edition

Copyright 2010 ASM