Scribd
Upload
6 views

FINA2230 Assignment 2 (24-25)

The document outlines Assignment 2 for FINA2230, due on February 20, 2025, focusing on various statistical concepts including measures of central tendency, properties of the Gamma function, variance calculations, and maximum likelihood estimation. It includes multiple questions that require the application of integration, probability distributions, and statistical inference methods. The assignment covers topics such as the chi distribution, sample variances, and loss data modeling.

Uploaded by

3C 01 蔡依涵
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
6 views

FINA2230 Assignment 2 (24-25)

The document outlines Assignment 2 for FINA2230, due on February 20, 2025, focusing on various statistical concepts including measures of central tendency, properties of the Gamma function, variance calculations, and maximum likelihood estimation. It includes multiple questions that require the application of integration, probability distributions, and statistical inference methods. The assignment covers topics such as the chi distribution, sample variances, and loss data modeling.

Uploaded by

3C 01 蔡依涵
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

24/25

FINA2230 Quantitative Methods for Actuarial Analysis II


Assignment 2

Due 20th Feb, 2025, 6:00 pm

1. Consider the U(0, 1) distribution and the following measures of central tendency
computed from a sample of size 3:
(a) sample mean
(b) sample median X(2)
(c) midrange Z  (X(1)  X(3))/2 defined in question 8 of assignment 1.
Find the mean and the variance of the three measures. Which one of the three
estimators of the population mean (1/2) has the smallest variance?

2. In this question, we explore properties of the Gamma function (y) which are useful

for FINA2230 and FINA3222. Recall that ( y )   u y 1e  u du for real y.
0

(a) Using integration by parts, show that for y  0, (y  1)  y(y). A direct result
of this, together with (1)  1, is that (y)  (y ‒ 1)! for integer y.
 ( y )
(b) Show that 0
u y 1e  ku du 
ky
. This is a handy formula that you should
memorize. For example, use it to show that Gam(, ) has kth moment
(  k )
k .
( )
(c) Let Y ~ 2(v)  Gam(  v/2,   2), and define U  Y1/2. U is said to follow the
chi distribution with df v.
Express E(U) and Var(U) in terms of v.
[Note: (v) has lots of applications in physics. For example, the speed of a molecule
from an ideal gas is (3).]

3. Let S2 be the variance of a random sample of size 16 from N(, 12).


(a) Find the mean and variance of S2.
(b) Find the mean and variance of S.

2
4. (a) The density of t(3) is f (t )  , t  .
 3(1  t 2 / 3) 2

1 1 t/ 3 t 
Show that its cdf is F(t )     tan 1 .
2  1  t / 3
2
3

(b) Let X and S be the sample mean and standard deviation of a sample of size 4
drawn from the N(1, 4) population. Find P( X  1  S).
24/25

5. Let S 2 be the variance of a random sample of size 16 from N(, 12), and T 2 be the
variance of a random sample of size 26 from N(2, 36).
Calculate P(2S 2 > T 2).

6. (a) Let c  0 and 0  p  c. Furthermore, let n and k (for 0  k < n) be positive


integers. Show that the function g(p)  pk(c ‒ p)n ‒ k is maximized at kc / n.
(b) The pmf of a population with a parameter  (with 0    1) is
k 0 1 2 3
pk /4 /4 2(1 ‒ )/3  (1 ‒ )/3
In a sample of size 36, you observe the following frequency table:
k 0 1 2 3
fk   13  4

Find  MLE by considering L() directly.

7. Consider a sample of claim counts per month of size 12. For the first 6 months, the
number of claims follows Poisson with mean . For the remaining 6 months, the
number of claims follows Poisson with mean 1.5.
(a) Show that l ( ; x)  12 x ln   15  c(x) for c(x) depending on x but not on .
(b) Hence, express the MLE for  in terms of the sampled values.
(c) Show that E( X )  1.25 . Hence, express the Fisher information in terms of .

8. You are given the following 12 loss data:


154, 215, 228, 265, 386, 490,
570, 709, 832, 925, 1056, 1411
You use the density f ( x; )  0.6 x 0.4 e  x for x  0 to model the data. Estimate 
0.6

(a) by MLE; (b) by matching the first moment.


[Note: For (b), use a suitable substitution and 2(b) to obtain the population mean.]

9. Consider the discrete population with probability mass function


p(0)  0.25, p(2)  0.60, and p(5)  0.15.
(a) Find the population variance 2.
The population variance is estimated using the biased sample variance T computed
from a random sample of size 2:
1
T  [( X 1  X ) 2  ( X 2  X ) 2 ] .
2
(b) By listing out all possible samples, find the distribution of T. Hence, find E(T).
(c) Find the mean square error of T as an estimator of the population variance:
E[(T ‒ 2)2].

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy