Large-N and Large-T Properties of Panel Data Estimators and the Hausman Test



Notice that since ∣∣Z⅞τ (E⅛^χ3jitEx3jit) ∣∣4 ≤ τ (r)4 G (¾)4 by Assumption
6(iv), we have sup,
∣∣Z⅞feT (A¾y.⅛fe,i - Ex3kti) ∣∣4 ≤ (ʃθɪ τ (r)4 dr) supiG(¾)4
for
к = 1,2, and 3. Therefore, by Lemma 10, we have

E KD3kτ (Ej^ix3k,i — Ex3k1i} K


(uniformly in i)



< ∞, for к = 1, 2, 3,


where τ3ι (r) = 0. So,

supE ∖∖D (Eτz.x3,i - Ex3ji) ∣∣4 < ∞.               (67)

Similarly, it follows that

sup sup E ∣∣Z⅛τ (E∏sx3 - E⅛)∣∣4 < ∞∙            (68)

N,T l<i<N


In addition, notice that

sup sup ∣∣Z⅛τ (Ex31i - Ex3)Il
N,Tl<i<N

sup sup


N,T l<i<N


(D33τ {Ex33tι — Ex33)
D3∙2τ (Ex321i — Ex32) — H32μg32i
D33τ {Ex33yι — Ex33) — H33μg33.


(69)


+ sup sup

N 3<i<N


0

-^32 At332

-^33 At333


By Assumption 6(v), as T → ∞,


sup sup


N 3<i<N


(D33τ (Ex33g. — Ex33)
D3,2τ
(¾2,i — Ex32) — H3g32 .

D33τ (Ex33ji — Ex33) — H33μg33 .


sup sup


D33τ {Ex33,i — Ex33j)            ''

D32τ (Ex32,i — Ex32j) — H32g32 i At3323J

D33τ (Ex33ιi — Ex33j) — H33 μg33 i — μ933 j) J


sup
i,j


f           D33τ (Ex33ji — Ex33j)

D32τ (Ex32,i — Ex32j) — H32 μg32 i At332i3

D33τ (Ex33,i — Ex33j) — H33g33 i — μ933 j


46




More intriguing information

1. The name is absent
2. Valuing Farm Financial Information
3. An alternative way to model merit good arguments
4. Evidence on the Determinants of Foreign Direct Investment: The Case of Three European Regions
5. Voluntary Teaming and Effort
6. Knowledge, Innovation and Agglomeration - regionalized multiple indicators and evidence from Brazil
7. THE INTERNATIONAL OUTLOOK FOR U.S. TOBACCO
8. Are class size differences related to pupils’ educational progress and classroom processes? Findings from the Institute of Education Class Size Study of children aged 5-7 Years
9. The name is absent
10. CAPACITAÇÃO GERENCIAL DE AGRICULTORES FAMILIARES: UMA PROPOSTA METODOLÓGICA DE EXTENSÃO RURAL