Download as pdf or txt
Download as pdf or txt
You are on page 1of 2

QF314901: Mathematical Statistics II

Assignment 03: Sufficient Statistic


Instructor: Chung-Han Hsieh (ch.hsieh@mx.nthu.edu.tw)
Teaching Assistant: Heng Tse Chou (hengtse.me@gmail.com)

Recalling that in class, we studied the Neyman’s factorization theorem; i.e.,


Theorem 3.1 (Neyman’s Factorization Theorem). Let X := (X1 , X2 , . . . , Xn ) be a random
sample from a population with pdf (or pmf ) f (x; θ) for θ ∈ Θ. Then T (X) is a sufficient statistic
for θ if and only if there exist two nonnegative functions g(·) and h(·) such that for all sample
points x and all parameters θ, the joint pdf of X can be factorized as

f (x; θ) = g(T (x); θ) · h(x)

where h(x) does not depend on θ.


Problem 3.1 (Standard Normal Sufficient Statistic for Variance). Let X := (X1 , . . . , Xn ) be
random sample from N (0, θ) with θ ∈ (0, ∞). Consider a statistic
n
X
T (X) := Xi2 .
i=1

Determine if T (X) is a sufficient statistic for θ or not.


Problem 3.2 (Multiple Order Statistics). Let X := (X1 , X2 , . . . , Xn ) be a continuous type
random sample from a population with common pdf (or pmf) f (x; θ) for unknown parameter
θ ∈ Θ. Define the T (X) := (Y1 , Y2 . . . , Yn ) with

Y1 := min Xi ≤ Y2 ≤ · · · ≤ Yn := max Xi
i i

be order statistics and the corresponding joint pdf of T (X); see Chapter 4 handout, is given by

fT (X) (y1 , . . . , yn ; θ) = n!f (y1 )f (y2 ) · · · f (yn )

for y1 ≤ · · · ≤ yn .
(i) Find the joint pdf of X.
(ii) Show that the ratio fT (X)f(y(x;θ)
1 ,...,yn ;θ)
= 1/n!.
(iii) According to part (ii), Is T (X) a sufficient statistic for θ ?
Problem 3.3 (Gamma Sufficient Statistic with Unknown Alpha). Let X := (X1 , . . . , Xn ) be a
random sample from a gamma(α, β) distributionwith pdf
1
f (x) = xα−1 e−x/β , x ∈ (0, ∞).
Γ(α)β α

Suppose that the parameter α := θ > 0 is unknown but β > 0 is known. Consider a statistic
n
X
T (X) = log(Xi ).
i=1

Is this T (X) sufficient for θ? Prove or disprove it.

1
Problem 3.4 (Useful Identities Involving Indicator Functions). Define an indicator function
1A (x) of a set A with (
1, x ∈ A
1A (x) :=
0, x ∈
/ A.
Qn
(i) Let xi ≥ 0 for
Qnall i. Show that i=1 1(0,θ) (xi ) = 1(0,θ) maxi (xi ).
(ii) Show that i=1 1(θ,∞) (xi ) = 1(θ,∞) (mini xi )
Problem 3.5 (Order Statistic as Sufficient Statistic). Let Y1 < Y2 < · · · < Yn be the order
statistics of a random sample X1 , . . . , Xn from the population with pdf

f (x; θ) = e−(x−θ) 1(θ,∞) (x)

with unknown parameter θ.


(i) Define Y1 := mini Xi . Determine if Y1 is a sufficient statistic of θ or not.
(ii) Define Yn := maxi Xi . Determine if Yn is a sufficient statistic of θ or not.
Problem 3.6 (Generalization of Uniform Sufficient Statistic). Let X := (X1 , . . . , Xn ) be random
sample of size n from the distribution

f (x; θ) := Q(θ)R(x) · 1(0,θ) (x)


Rθ 1
with 0 R(x)dx = Q(θ) where θ ∈ (0, ∞) is unknown parameter and 1A (x) is the indicator
function of a set A. To estimate θ, we form a statistic

T (X) := max Xi .
1≤i≤n

Show that T is sufficient for θ.

You might also like