The WLS estimator is consistent
Problem
- Setup
- Random sample \(\left( y_i,x_i \right)\) for \(i=1, \ldots ,n\)
- Correctly specified linear regression model, \(y=Xβ+ε\) where \(E\left( ε \mid X \right)=0\) .
- \(Var\left( ε_i \mid x_i \right)=σ^2σ_i^2\) where \(σ^2\) is unknown and \(σ_i^2\) is known or in matrix form, \(Var\left( ε \mid X \right)=σ^2Ψ\) where \(Ψ=diag\left( σ_1^2, \ldots ,σ_n^2 \right)\) .
- \(b_{WLS}={\left( X'Ψ^{-1}X \right)}^{-1}X'Ψ^{-1}y\)
Define
\[Σ=E\left( \frac{x_ix'_i}{σ_i^2} \right)\]
Assume that \(Σ\) is invertible. Show that \(b_{WLS}\) is consistent.
Solution
The solution is very similar to proving consistency of the OLS estimator.
Start from \(b_{WLS}=β+{\left( X'Ψ^{-1}X \right)}^{-1}X'Ψ^{-1}ε\) .
\[plim b_{WLS}=β+plim {\left( X'Ψ^{-1}X \right)}^{-1}X'Ψ^{-1}ε=\]
\[β+plim {\left( n^{-1}X'Ψ^{-1}X \right)}^{-1}⋅plim n^{-1}X'Ψ^{-1}ε\]
We have
\[n^{-1}X'Ψ^{-1}X=n^{-1}\sum_{i=1}^{n}{ \frac{x_ix'_i}{σ_i^2} }\]
and
\[plim {\left( n^{-1}X'Ψ^{-1}X \right)}^{-1}=Σ\]
For the next term,
\[E\left( n^{-1}X'Ψ^{-1}ε \right)=E\left( E\left( n^{-1}X'Ψ^{-1}ε \mid X \right) \right)=0\]
and
\[Var\left( n^{-1}X'Ψ^{-1}ε \right)=E\left( Var\left( n^{-1}X'Ψ^{-1}ε \mid X \right) \right)=n^{-2}σ^2E\left( X'Ψ^{-1}X \right)=n^{-1}σ^2Σ→0\]
so
\[plim n^{-1}X'Ψ^{-1}ε=0\]
Combining,
\[plim b_{WLS}=β+Σ^{-1}⋅0=β\]