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



as (N, T → ∞). Now, recall that

d2T

D3T


diag (D21τ,D22τ) = diag Ik2,, √T⅛2} ;

diag (D31τ,D32τ,D33τ) = diag TIlk3γ, VTlk32, (Tmh)ft) ,

where 0 mh,33 < ɪ for all h = 1, ...,k33. Therefore, as (N, T → ∞) ,

/i21,21

121,22

I21,31

I21,32

I21,33

0

0

0

0

0

I0
i21,22

I22,22

I22,31

I22,32

I22,33

0

γ22,22

Γ22,31

Γ22,32

0

I0
i21,31

I0

i 22,31

I31,31

I31,32

I31,32

P

0

Γ'

γ 22,31

Γ31,31

Γ31,32

0

I0
i21,32

I0

I22,32

I0

I31,32

I32,32

I32,33

0

Γ0

Γ22,32

Γ0

Γ31,32

Γ32,32

0

121,33

I0

I22,33

I0

I31,33

I0

I32,33

I33,33

0

0

0

0

0

Finally, by the Cauchy-Schwarz inequality, the off-diagonal block component

I1,21 I1,22 I1,31   I1,32 I1,33 ) p 0,

as (N, T → ∞).

Proof of (49): Recallthat

Etiuji = (Ep^E^EF-zi x3,i ,z'i)' ;

E*w = (Ex1, Ex2, EfzX3, z,).

Write

I2,i,NT

/                        D1τ (EX1,l - EX1)                        

EX21,l - EX21

0

=    D31τ ((EFzi X31,i - EX31,i) - (EFz5⅛1 - EX31) + (EX31,i - EX31))

D32τ ((EFzi X32,i - EX32,i) - (EfzX32 - EX32) + (EX32,i - EX32))
D33τ ((EriX33,i - EX33,i) - (Efzx33 - EX33) + (EX33,i - EX33))

                             Zi - z                              /

To obtain the required result, we use Lemma 12. By Assumption 6, as T → ∞,
we have

I2,i,NT I2,i,N a.s. uniformly in i,                   (52)

where

/                          H1Θ 1,i                          

Θ 21,i

0

I -                             0

I21,i,N =                           1                              г            1              1    .

H32 Lg32,i (zi) - N ∑i g32,i (zi)J + H32 μg32,i - N ∑i μg32,i

H33 [g33,i (zi) - N Pi g33,i (zi)] + H33 μg33,i - N Pi μg32,i

V                      zi - N Pi zi                      /

40



More intriguing information

1. Non-farm businesses local economic integration level: the case of six Portuguese small and medium-sized Markettowns• - a sector approach
2. Skills, Partnerships and Tenancy in Sri Lankan Rice Farms
3. The name is absent
4. The name is absent
5. Volunteering and the Strategic Value of Ignorance
6. The InnoRegio-program: a new way to promote regional innovation networks - empirical results of the complementary research -
7. Can a Robot Hear Music? Can a Robot Dance? Can a Robot Tell What it Knows or Intends to Do? Can it Feel Pride or Shame in Company?
8. Eigentumsrechtliche Dezentralisierung und institutioneller Wettbewerb
9. The Global Dimension to Fiscal Sustainability
10. The name is absent