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



Then, under Assumption 11, as (N, T →∞), the followings hold.

(a) NN Pi Dtwiw0DT p ξ

(b) supN tsup1iN E \\DTwik4 < M, for some constant M < .

(c) √N Pi DTwi N (≡λ, σU≡0 .

(d) N Pi DTW^iVi p 0.

Proof

Part (a)

To find the joint limit of

N X DTwJiwJ0DT =
i


1

UD (wji
i


w) (wi w)' DT,


we define

Ew = Ex^Ex'^EFiχ3,i,zi)' ;

E*W = (Ex01,Ex'2,EFzx'3,z').

With this notation, we write

1

NλJh (wi

i


w) (Wi


w)' Dt


= NXDt [(wui E*iw^i) + (E*iWi E*wj) + (E*w wu)]

i

× [(w7i Evi) + (EiWi E*w) + (E*w w)]' Dt

-1X

N
i


(I1,i,N T+I2,i,NT+I3,N T)(I1,i,N T+I2,i,N T+I3,N T)' , say.


(47)


We complete the proof by showing the following:

N X I1,i,NTI1 ,i,NT
i

p Ξ1;

(48)

N X I2,i,NTI2,i,NT
i

p Ξ2;

(49)

I3,N T p 0;

and

(50)

NX I1,i,NTI2,i,NT, nX I1,i,NTI3,NT ,

nX I2,i,NTI3,NTp 0.
i

(51)

Proof of (48): Write

-1 X I1,i,NTI1 ,i,NT
i

38



More intriguing information

1. Lending to Agribusinesses in Zambia
2. The name is absent
3. A Note on Productivity Change in European Co-operative Banks: The Luenberger Indicator Approach
4. The name is absent
5. Synchronisation and Differentiation: Two Stages of Coordinative Structure
6. The name is absent
7. The name is absent
8. The use of formal education in Denmark 1980-1992
9. Sectoral Energy- and Labour-Productivity Convergence
10. Beyond Networks? A brief response to ‘Which networks matter in education governance?’