Scribd
Upload
89 views

Advanced Econometric Methods I: Problem Set 3: Geert Mesters October 13, 2020

This problem set for an advanced econometrics course includes 4 questions about statistical hypothesis testing. Students need to submit their solutions by email before the next teaching assistant session on October 19th. The questions cover topics like: finding the variance of an unbiased error term estimator; determining the uniformly most powerful test between two normal distributions; properties of power functions; and finding the uniformly most powerful test for uniform distributions.

Uploaded by

Kadir
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
89 views

Advanced Econometric Methods I: Problem Set 3: Geert Mesters October 13, 2020

This problem set for an advanced econometrics course includes 4 questions about statistical hypothesis testing. Students need to submit their solutions by email before the next teaching assistant session on October 19th. The questions cover topics like: finding the variance of an unbiased error term estimator; determining the uniformly most powerful test between two normal distributions; properties of power functions; and finding the uniformly most powerful test for uniform distributions.

Uploaded by

Kadir
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 1

Advanced Econometric Methods I: Problem set 3

Geert Mesters
October 13, 2020

This problem set includes 4 questions. Please explain your answers clearly and state
additional assumptions clearly if you use them. You may work in small groups (max 4) and
you hand in 1 assignment per group. Your solutions need to be emailed to Vladislav
([email protected]) before the next TA-session 19-10-2019 (Monday).
Good luck !!!
• Before answering the questions below it is recommended to read the relevant pages in
Hayashi and Casella & Berger. See the detailed reading list on google classroom.

(i) In the normal linear regression model (e.g. assumptions 1-5 of Hayashi Chapter 1 hold),
let s2 be the unbiased estimator of the error variance σ 2 . Show that the variance of s2
is strictly larger than the Cramr-Rao Lower Bound for σ 2 .

(ii) Given a random sample from a N (0, σ 2 ) distribution, find the UMP test for H0 : σ = σ0
versus σ = σ1 (let σ0 < σ1 ). What is the corresponding critical region?

(iii) Suppose X is one observation from a population with β(1, θ) distribution.

(a) For testing H0 : θ ≤ 0.5 versus H1 : θ > 0.5 sketch the power function of the test
that rejects H0 if X > 0.5.
(b) Find the most powerful test of level α for H0 : θ = 1 against H1 : θ = 2
(c) Is there UMP test for H0 : θ ≤ 1 versus H1 : θ > 1? If so find it. If not prove
that.

(iv) Let X1 , . . . , Xn be a random sample from the uniform(θ,1 + θ) distribution. To test


H0 : θ = 0 versus H1 : θ > 0 use the test

reject if max Xi ≥ 1 pr min Xi ≥ k

(a) Show that (min Xi , max Xi ) is a sufficient statistic for θ.


(b) Determine k such that the test has size α
(c) Find an expression for the power function of this test (using part b).
(d) Show that the test is UMP

You might also like