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



By definition,


N1 X EFzi kRiTk
i

sup EFzi kRiT k ≤ sup EFzi kQiTk1{kQiTk >M}.       (39)

Also,



N X(PiT - EFzi PiT )
i


(⅛ X EFzi ° PiT - EFZi PiT ° 22

≤√⅛ μsiuτp EFzi kPiT k2)2


(40)

Choose M = N a, where 0 < a < 2. Then, (39) , (40) 0 α.s.. Consequently,
we have (38).
¥

Lemma 15 Suppose that Assumptions 1-8 hold. Let Fz = σ (z1, ..., zN , ...) .
For a generic constant M that is independent of N and T, and for some Fz -
measurable function Mz , the followings hold:

(a) suPi,τ TT Pt (xι,it - Exι,it) °4 M;

(b) suPi,T T Pt (χ2,it - Eχ2,it) ∣∣4 M;

(c) suPi,T 1T Pt (χ3,it - EFziχ3,i0 °F 4 < Mz, a.s.;

(d) suPi,T ° T Pt (χ3,it - EFzi χз,it¢ ∣∣4 M.

Proof

We here use q to denote the real number used in Assumptions 1-6, which is
strictly greater than 1.

Part (a)

Note that

sup
i,T


T X (x1,it - Ex1,it)

t                         °4


sup kx1,it - Ex1,itk4

i,T

sup kx1,it - Ex1,itk4q = κx2 ,
i,T                      q

where the first inequality holds by Minkowski’s inequality, the second inequality
holds by Liapunov’s inequality, and the last inequality holds by Assumption 2.
Choose M = κ
x2 . ¥

Part (b)

34



More intriguing information

1. Auction Design without Commitment
2. Long-Term Capital Movements
3. How we might be able to understand the brain
4. The name is absent
5. On the origin of the cumulative semantic inhibition effect
6. Citizenship
7. Contribution of Economics to Design of Sustainable Cattle Breeding Programs in Eastern Africa: A Choice Experiment Approach
8. Evolution of cognitive function via redeployment of brain areas
9. The Impact of Financial Openness on Economic Integration: Evidence from the Europe and the Cis
10. Spatial agglomeration and business groups: new evidence from Italian industrial districts