Académique Documents
Professionnel Documents
Culture Documents
October, 2022
yi = xi′ β + ϵi
Suppose that
yt = β1 + β2 xt + β3 yt−1 + ϵt
and that ϵt = ρϵt−1 + vt where vt is white noite
Now,
and thus
ρσ 2
cov (yt−1 , ϵt ) =
1 − ρβ3
yt = β1 + β2 xt + ϵt
xt = yt + zt
β1 β2 zt ϵt
yt = + +
1 − β2 1 − β2 1 − β2
yi = xi∗ ′ β + ϵi
cov [xi , yi ]
βb =
var [xi ]
cov [xi , wi ]
plimβb = β + γ
var [xi ]
| {z }
Omitted Variable Bias
yi = β1 + β2 xi + ϵi
yi = β1 + β2 xi + ϵi
to obtain
cov (yi , zi ) = β2 cov (xi , zi ) + cov (ϵi , zi )
and thus
cov (yi , zi )
β2 =
cov (xi , zi )
E(ϵi zi ) = 0
V(
b β b 2 (Z′ X)−1 Z′ Z(X′ Z)−1
b )=σ
IV
with
N
1 X
σ2
b = (yi − xi′ βbIV )
N i=1
since the results are valid asymptotically it does not matter whether we
correct for degrees of freedom
standard errors of IV estimators are typically quite high when compared
to OLS - this is usually due to low correlation between the instrument
and the regressor
yi = xi′ β + ϵi
by
N
1 X
(yi − xi′ β)zi = 0
N i=1
N
!−1
opt 1 X
WN = zi z′
N i=1 i
V(
b β b 2 (X′ Z(Z′ Z)−1 Z′ X)−1
b )=σ
IV
where
N
1 X
b2 =
σ (yi − xi′ βbIV )
N i=1
βbIV = (X b −1 X
b ′ X) b ′Y
where
b = Z(Z′ Z)−1 Z′ X
X
′ −1
bOLS − bIV VarIV − VarOLS bOLS − bIV ∼ χ2k
′ −1
′
X X X
N (1/N) εbi zi (1/N) zi zi (1/N) εbi zi ∼ χ2k