Beruflich Dokumente
Kultur Dokumente
i
i
E Y y N
E Y y N
= = =
= =
i
y N y = =
and
Eq. 10.11
i
y N y = =
( )
2
2 2 2
2 2 2
2
i i i
y y y y Ny
y
N N N
= = = =
% Eq. 10.12
In the linear regression model y =
1
+
2
x + e, we
usually assume:
10.3.2
Method of Moments
Estimation in the
Simple Linear
Regression Model
10.3
Estimators Based on
the Method of
Moments
Eq. 10.13
( ) ( )
1 2
0 0
i i i
E e E y x = =
If x is fixed, or random but not correlated with
e, then:
Eq. 10.13
Eq. 10.14
( ) ( )
1 2
0 0
i i i
E e E y x = =
( ) ( )
1 2
0 0 E xe E x y x = ( =
We have two equations in two unknowns:
10.3.2
Method of Moments
Estimation in the
Simple Linear
Regression Model
10.3
Estimators Based on
the Method of
Moments
1
Eq. 10.15
( )
( )
1 2
1 2
1
0
1
0
i i
i i i
y b b x
N
x y b b x
N
=
=
Eq. 10.17
Solving these equations leads us to method of
moments estimators, which are usually called the
instrumental variable (IV) estimators:
10.3.3
Instrumental
Variables Estimation
in the Simple Linear
Regression Model
10.3
Estimators Based on
the Method of
Moments
Eq. 10.18
( )( )
( )( )
2
1 2
i i
i i i i
i i i i i i
z z y y N z y z y
N z x z x z z x x
y x
= =
=
These new estimators have the following
properties:
They are consistent, if z is exogenous, with
E(ze) = 0
In large samples the instrumental variable
10.3.3
Instrumental
Variables Estimation
in the Simple Linear
Regression Model
10.3
Estimators Based on
the Method of
Moments
In large samples the instrumental variable
estimators have approximate normal
distributions
In the simple regression model:
Eq. 10.19
( )
2
2 2
2
2
~ ,
zx i
N
r x x
| |
|
|
2
i i
IV
y x
N
=
var
zx
zx i
b
r
r x x
= =
K K
y x x e = + + + + L
are the instrumental variables (IV) estimators
Because they can be obtained by two least
squares regressions, they are also popularly
known as the two-stage least squares (2SLS)
estimators
We will refer to them as IV or 2SLS or
IV/2SLS estimators
In the simple regression, if x is endogenous and
we have L instruments:
10.3.4a
Using Surplus
Instruments in
Simple Regression
10.3
Estimators Based on
the Method of
Moments
1 1 1
L L
x z z = + + + L
The two sample moment conditions are:
( )
( )
1 2
1 2
1
0
1
0
i i
i i i
y x
N
x y x
N
=
=
x x y y
x x =
( )
( )
( )
( )
( )( )
( )( )
2
1 2
i i
i i
i i
i i
x x y y
x x y y
x x x x
x x x x
y x
= =
=
( ) ( ) ( ) ( ) ( )
2
ln 0.0481 0.0614 0.0442 0.0009
se 0.4003 0.0314 0.0134 0.0004
WAGE EDUC EXPER EXPER = + +
To compare, the OLS estimates of the log-linear wage
equation are:
( )
( ) ( ) ( ) ( ) ( )
2
ln 0.1982 0.0493 0.0449 0.0009
se 0.4729 0.0374 0.0136 0.0004
WAGE EDUC EXPER EXPER = + +
Estimation Issue 2: Identification
The multiple regression model, including all K
variables, is:
G exogenous variables B endogenous variables
1 2 2 1 1
G exogenous variables B endogenous variables
G G G G K K
y x x x x e
+ +
= + + + + + +
L L
Think of G = Good explanatory variables, B = Bad
explanatory variables and L = Lucky instrumental
variables
It is a necessary condition for IV estimation that
L B
If L = B then there are just enough instrumental
variables to carry out IV estimation
10.3.8
Instrumental
Variables Estimation
in a General Model
10.3
Estimators Based on
the Method of
Moments
variables to carry out IV estimation
The model parameters are said to just identified
or exactly identified in this case
The term identified is used to indicate that the
model parameters can be consistently estimated
If L > B then we have more instruments than are
necessary for IV estimation, and the model is said
to be overidentified