Intermediate Statistics - Homework 0: Piecewise Exponential Distribution - Prof. Jianhua H, Assignments of Statistics

The instructions and problems for homework 0 of the intermediate theory of statistics course (stat610) at the university of x. Students are required to solve problem 1, which deals with a piecewise exponential distribution. The problem asks students to show that the given density function is a probability density function, find the probability and expected value for a specific event, and find the maximum likelihood estimator for the unknown parameters. The document also includes a solution by student b.

Typology: Assignments

Pre 2010

Uploaded on 02/10/2009

koofers-user-jbx
koofers-user-jbx 🇺🇸

10 documents

1 / 1

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
Stat610: Intermediate Theory of Statistics
Homework 0 (8/29/06)
Instructor: Jianhua Huang Editor: Student A
TA: Seokho Lee
Send your solution (latex file) of the problem to the assigned editor of the week. After combining the files
and editing, the assigned editor should send the file to the TA (Seokho Lee, [email protected]u.edu).
Thanks!
Jianhua
Problem 1.
A random variable Xhas piecewise exponential distribution with parameters λ1,...,λmand partition
t1,...,tmif its density is
f(x) = kjλ1
je(xtj1)j, x (tj1, tj], j = 1,...,m
where t0= 0, tm=,k1= 1 and
kj= exp
j1
X
l=1
(tltl1)l, j = 2,...,m.
a) Show that f(x) is indeed a probability density function, that is,
R
0f(x)dx = 1.
b) Show that P(X > tj1) = kjand
E[(Xitj1)I(tj1Xitj)] = λj(tjtj+1)(tjtj+1)kj+1 .
c) Suppose X1,...,Xnis a random sample from above piecewise exponential distribution and λ1,...,λmare
unknown parameters that need to be estimated. Give the closed-from expression of the MLE of λ1,...,λm.
d) Show that the MLE is consistent.
Solution (by Student B).
1

Partial preview of the text

Download Intermediate Statistics - Homework 0: Piecewise Exponential Distribution - Prof. Jianhua H and more Assignments Statistics in PDF only on Docsity!

Stat610: Intermediate Theory of Statistics

Homework 0 (8/29/06)

Instructor: Jianhua Huang Editor: Student A TA: Seokho Lee

Send your solution (latex file) of the problem to the assigned editor of the week. After combining the files and editing, the assigned editor should send the file to the TA (Seokho Lee, [email protected]).

Thanks!

Jianhua

Problem 1.

A random variable X has piecewise exponential distribution with parameters λ 1 ,... , λm and partition t 1 ,... , tm if its density is

f (x) = kj λ− j 1 e−(x−tj−^1 )/λj^ , x ∈ (tj− 1 , tj ], j = 1,... , m

where t 0 = 0, tm = ∞, k 1 = 1 and

kj = exp

∑^ j−^1

l=

(tl − tl− 1 )/λl

, j = 2,... , m.

a) Show that∫ f (x) is indeed a probability density function, that is, ∞ 0 f^ (x)^ dx^ = 1.

b) Show that P (X > tj− 1 ) = kj and

E[(Xi − tj− 1 )I(tj− 1 ≤ Xi ≤ tj )] = λj (tj − tj+1) − (tj − tj+1)kj+1.

c) Suppose X 1 ,... , Xn is a random sample from above piecewise exponential distribution and λ 1 ,... , λm are unknown parameters that need to be estimated. Give the closed-from expression of the MLE of λ 1 ,... , λm.

d) Show that the MLE is consistent.

Solution (by Student B).