yit=x′itβ1+yi(t−1)β2+αi+εit
As in the conventional linear panel data model the time-invariant unobserved component is related to the regressors but the strategy of taking first differences does not work because of an endogeneity problem. In particular:
Δyit=Δx′itβ1+Δyi(t−1)+Δεit
As in the conventional linear panel data model the time-invariant unobserved component is related to the regressors but the strategy of taking first differences does not work because of an endogeneity problem. In particular:
Δyit=Δx′itβ1+Δyi(t−1)+Δεit
Comment