This preview shows pages 1–2. Sign up to view the full content.
D. Steigerwald
Department of Economics
Economics 140A
Exercise 5
1.
(A difficult question.
Answer for extra credit.)
We will investigate the following claim
(which appears in nearly this form in Studenmund's undergraduate econometrics text)
If an equation that contains a lagged dependent variable as a regressor has a serially
correlated error term, then OLS estimates of the coefficients will be inconsistent.
To investigate, we define a regression model in a slightly different way.
Let
t
Y
be a
strictly stationary and ergodic random variable that is observed into the infinite past.
Assume that
0
t
EY
and
2
t
EY
.
Define
t
t
t
Y
Var
Y
Y
Cov
,
1
and
1
t
t
t
Y
Y
U
so that we can write
t
t
t
U
Y
Y
1
for
t=1,…,n
.
a.
Show that
0
t
EU
and
0
,
1
t
t
U
Y
Cov
.
b.
Show that the OLS estimator
B
from a regression of
t
Y
on
1
t
Y
, for
t=1,…,n
, is
consistent for
β
d.
What do you conclude about the validity of the quotation?
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
This is the end of the preview. Sign up
to
access the rest of the document.
This note was uploaded on 09/04/2011 for the course ECON 140a taught by Professor Staff during the Fall '08 term at UCSB.
 Fall '08
 Staff
 Economics

Click to edit the document details