Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
090507
1/22
Panel Data
or
Potential gains
take heterogenity into account, get individual-speci…c
estimates
especially suitable to study dynamics of change
study more sophisticated behavioral models
minimize bias due to aggregation
Balanced/unbalanced
The intercept
The slope coe¢ cients
For N = 4 and T = 20
Cautions
β1i = β1 + εi i = 1, ..., N
9
εi N 0, σ2ε >
>
>
>
>
>
>
E ( εi εj ) = 0 for i 6= j >
>
>
>
=
uit N 0, σ2u
>
>
>
>
E (uit uis ) = E (uit uit ) = E (uit ujs ) = 0 for i 6= j t 6= s >
>
>
>
>
>
>
;
E (εi uit ) = 0
8
> E (wit ) = 0
>
>
>
<
) Var (wit ) = σ2ε + σ2u
>
>
>
>
: σ2ε
Corr (wit , wis ) = σ2ε +σ2u
depends on
E (εi Xi ) = 0?
If yes: random e¤ects, if no: …xed e¤ects
the relation between T and N
- for large T and small N not a big di¤erence
- for small T and large N random e¤ects estimators are more
e¢ cient than …xed e¤ects (if the assumptions hold)