A Principal Components Approach to Cross-Section Dependence in Panels

N =30,T

=25, γi

= 1, ρ_di

=0, ρ_z

=0, σ²_di =

1

A) Regression I

Mean

SSD

Min.

Max.

Skew.

Kurt-3

MG________________________________

ʌ
^

1.5070

.0766

1.2645

1.7224

-.2967

.0069

ʌ

SEι(θ)

.0305

.0048

.0171

.0479

.2111

.0868

FE_________________________________________

ʌ
^

1.4966

.0754

1.2460

1.7053

-.3233

.0326

ʌ

SEι(θ)

.0322

.0020

.0259

.0378

-.0169

-.1542

ʌ

SE2(θ)

.0567

.0035

.0465

.0659

.0043

-.0825

ʌ

SE3(θ)

.0716

.0137

.0410

.1318

.8815

1.2679

POLS_______________________________________

ʌ
^

1.4869

.0751

1.2406

1.6989

-.3207

.0052

ʌ

SE1(θ)

.0317

.0019

.0256

.0375

-.0124

-.1353

ʌ

SE2(θ)

.0567

.0034

.0461

.0656

.0082

-.0888

ʌ

SE3(θ)

.0717

.0132

.0366

.1296

.4597

.2201

B) Regression II

Mean

SSD

Min.

Max.

Skew.

Kurt-3

MG_______________________________

ʌ
b

1.1384

.0734

.9472

1.4178

.3728

.2719

ʌ

SE1(b)

.0374

.0052

.0201

.0539

.1556

-.1378

FE_________________________________________

ʌ
b

1.1279

.0709

.9565

1.3694

.3425

.2523

ʌ

SE1(b)

.0341

.0015

.0278

.0378

-.0736

.1013

ʌ

SE2(b)

.0355

.0024

.0297

.0488

.6921

1.3834

ʌ

SE3(b)

.0406

.0076

.0246

.0801

.8646

1.8745

POLS_______________________________________

ʌ
b

1.1243

.0689

.9539

1.3793

.3591

.2305

ʌ

SEi(b)

.0335

.0016

.0279

.0386

-.1084

.1745

ʌ

SE2(b)

.0355

.0024

.0299

.0492

.7604

1.8043

ʌ

SE3(b)

.0392

.0076

.0220

.0642

.5005

.2672