Econ 483. Computer Exercise 4
Data are on the Catalyst Discussion Board (check myUW).
Name:
Exercise A.
endogenous explanatory variables with known data generating process
1. Download dgpendog.dta and open it by using
use
command. How the sample was generated is as
follows. The original data had 6 variables:
V
1
,
V
2
,
V
3
,
V
4
,
V
5
,
V
6
. The
V
variables are realized numbers
from independent and identically distributed normal with expectation zero and variance one. In other
words,
E
[
V
m
] = 0
and
V ar
[
V
m
] = 1
for
m
= 1
;
2
; :::;
6
. Also,
Cov
[
V
l
; V
m
] = 0
for
l
6
=
m
.
The current data has
X
1
=
V
1
+
V
5
,
X
2
=
V
2
+
V
5
+
V
6
, and
U
=
V
6
. In addition,
Z
1
=
V
2
+
V
3
and
Z
2
=
V
2
+
V
4
. Finally, Generate
Y
=
&
0
+
1
X
1
+
2
X
2
+
U
where
&
0
= 2
:
8
,
1
= 4
:
2
, and
2
= 3
:
2
.
(1) Calculate
Cov
[
X
1
; U
]
and
Cov
[
X
2
; U
]
.
(2) Calculate
Cov
[
X
2
; Z
1
]
and
Cov
[
X
2
; Z
2
]
. Also, calculate
Cov
[
U; Z
1
]
and
Cov
[
U; Z
2
]
.
(3) We estimate
Y
i
=
±
0
+
±
1
X
1
i
+
±
2
X
2
i
+
U
i
. In this regression model, do you have any endogenous
(right hand side) variables? Do you have any exogenous (right hand side) variables?
(4) Estimate the above model and test
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.
 Spring '08
 Staff
 Macroeconomics, Regression Analysis, SMSA

Click to edit the document details