ECON | 5370 | Advanced Econometric Theor...
ECON | 5370 | Advanced Econometric Theory III | lecture_notes |Simultaneous Equation Models 1 Introduction Many economic problems involve the interaction of multiple endogenous variables within a system of equa- tions. Estimating the parameters of such as system is typically not as simple as doing OLS equation-by- equation. Issues such as identication (whether the parameters are even estimable) and endogenei
Showing 8-9 out of 12
ECON | 5370 | Advanced Econometric Theory III | le...
ECON | 5370 | Advanced Econometric Theory III | lecture_notes |Simultaneous Equation Models 1 Introduction Many economic problems involve the interaction of multiple endogenous variables within a system of equa- tions. Estimating the parameters of such as system is typically not as simple as doing OLS equation-by- equation. Issues such as identication (whether the parameters are even estimable) and endogenei
ECON | 5370 | Advanced Econometric ...
ECON | 5370 | Advanced Econometric Theory III | lecture_notes |Simultaneous Equation Models 1 Introduction Many economic problems involve the interaction of multiple endogenous variables within a system of equa- tions. Estimating the parameters of such as system is typically not as simple as doing OLS equation-by- equation. Issues such as identication (whether the parameters are even estimable) and endogenei
Page 8
plim
(
1
T
W
0
i
Z
i
)=
wz
, a nite invertible matrix
plim
(
1
T
W
0
i
W
i
)=
ww
, a nite positive-denite matrix
plim
(
1
T
W
0
i
i
)=0
.
Since plim
(
Z
0
i
i
)
6
=0
, we can instead examine
plim
(
1
T
W
0
i
y
i
)
=
plim
(
1
T
W
0
i
Z
i
)
+
plim
(
1
T
W
0
i
i
)
=
plim
(
1
T
W
0
i
Z
i
)
.
Naturally, the
instrumental
variable estimator is
^
IV
i
=(
W
0
i
Z
i
)
1
(
W
0
i
y
i
)
which is consistent and has asymptotic variance-covariance matrix
asy:var:
(
^
IV
i
)=
ii
T
[
1
wz
ww
1
zw
]
:
This can be estimated by
est:asy:var:
(
^
IV
i
)=^
ii
(
W
0
i
Z
i
)
1
W
0
i
W
i
(
Z
0
i
W
i
)
1
and
^
ii
=
1
T
(
y
i
Z
i
^
IV
i
)
0
(
y
i
Z
i
^
IV
i
)
:
A degrees of freedom correction is optional.
Notice that ILS is a special case of IV estimation for an exactly
identied equation where
W
i
=
X
.
4.4
Two-Stage Least Squares (2SLS)
When an equation in the system is overidentied (i.e.,
rows
(
X
0
Z
i
)
>
cols
(
X
0
Z
i
)
), a convenient and intuitive
IV estimator is the
two-stage
least
squares estimator.
The 2SLS estimator works as follows:
Stage
#1. Regress
Y
i
on
X
and form
^
Y
i
=
X
^
OLS
.
Stage
#2. Estimate
i
by an OLS regression of
y
i
on
^
Y
i
and
X
i
.
More formally, let
^
Z
i
=(
^
Y
i
;X
i
)
.
The 2SLS estimator is given by
^
2
SLS
i
=(
^
Z
0
i
^
Z
i
)
1
(
^
Z
0
i
y
i
)
8


Page 9
where the asymptotic variance-covariance matrix for
^
2
SLS
i
can be estimated consistently by
est:asy:var
(
^
2
SLS
i
)=^
ii
(
^
Z
0
i
^
Z
i
)
1
and
^
ii
=
1
T
(
y
i
Z
i
^
2
SLS
i
)
0
(
y
i
Z
i
^
2
SLS
i
)
.
4.5
Limited-Information Maximum Likelihood (LIML)
Limited-information
maximum
likelihood estimation refers to ML estimation of a single equation in the
system.
For example, if we assume normally distributed errors, then we can form the joint probability
distribution function of
(
y
i
;Y
i
)
and maximize it by choosing
i
and the appropriate elements of
.
Since
the LIML estimator is more complex but asymptotically equivalent to the 2SLS estimator, it is not widely
used.
Note.
When the
i
th
equation is exactly identied,
^
ILS
i
=
^
IV
i
=
^
2
SLS
i
=
^
LIML
i
.
5
Full-Information Estimation
The ve estimators mentioned above are not fully e¢ cient because they ignore cross-equation relationships
between error terms and any omitted endogenous variables.
We consider two fully e¢ cient estimators below.
5.1
Three-Stage Least Squares (3SLS)
Begin by writing the system (1) as
2
6
6
6
6
6
6
6
4
y
1
y
2
.
.
.
y
M
3
7
7
7
7
7
7
7
5
=
2
6
6
6
6
6
6
6
4
Z
1
0

0
0
Z
2
0
.
.
.
.
.
.
.
.
.
0
0

Z
M
3
7
7
7
7
7
7
7
5
2
6
6
6
6
6
6
6
4
1
2
.
.
.
M
3
7
7
7
7
7
7
7
5
+
2
6
6
6
6
6
6
6
4
1
2
.
.
.
M
3
7
7
7
7
7
7
7
5
)
y
=
Z
+
where
=
~
E
and
E
(

0
)=
I
T
.
Then, applying the principle from SUR estimation, the fully e¢ cient estimator is
^
=(
W
0
(
1
I
)
Z
)
1
(
W
0
(
1
I
)
y
)
9


Ace your assessments! Get Better Grades
Browse thousands of Study Materials & Solutions from your Favorite Schools
University of Wyoming
University_of_Wyoming
School:
Advanced_Econometric_Theory_III
Course:
Great resource for chem class. Had all the past labs and assignments
Leland P.
Santa Clara University
Introducing Study Plan
Using AI Tools to Help you understand and remember your course concepts better and faster than any other resource.
Find the best videos to learn every concept in that course from Youtube and Tiktok without searching.
Save All Relavent Videos & Materials and access anytime and anywhere
Prepare Smart and Guarantee better grades

Students also viewed documents