A Principal Components Approach to Cross-Section Dependence in Panels



Table 1A. Monte Carlo simulation results

N =30,T

= 300, γi

= 1, ρdi

=0, ρz

=0, σ2di

=1

A) Regression I

Mean

SSD

Min.

Max.

Skew.

Kurt-3

MG estimator

ʌ
^

1.4983

.0217

1.4243

1.5584

.0544

.0215

ʌ

SEι(θ)

.0083

.0011

.0056

.0116

.1819

-.2505

FE estimator

ʌ
^

1.4975

.0216

1.4244

1.5579

.0593

.0299

ʌ

SEι(θ)

.0092

.0002

.0087

.0096

-.1818

.0738

ʌ

SE2(θ)

.0169

.0003

.0161

.0179

.2264

.2835

ʌ

SE3(θ)

.0222

.0014

.0189

.0291

.7026

1.3561

POLS estimator

ʌ
^

1.4967

.0216

1.4232

1.5573

.0589

.0382

ʌ

SE1(θ)

.0092

.0002

.0087

.0096

-.1829

.0829

ʌ

SE2(b)

.0169

.0003

.0161

.0179

.2299

.3072

ʌ

SE3(θ)

.0221

.0015

.0180

.0277

.5705

.6412

B) Regression II

Mean

SSD

Min.

Max.

Skew.

Kurt-3

MG_______________________________

ʌ
b

1.0987

.0165

1.0503

1.1464

.2764

-.1859

ʌ

SE1(b)

.0100

.0013

.0063

.0137

.1158

-.0758

FE_________________________________________

ʌ
b

1.0980

.0163

1.0517

1.1459

.2769

-.1551

ʌ

SE1(b)

.0097

.0001

.0094

.0101

.0156

-.0846

ʌ

SE2(b)

.0098

.0001

.0095

.0103

.2678

.1741

ʌ

SE3(b)

.0109

.0006

.0092

.0133

.2375

.3445

POLS_______________________________________

ʌ
b

1.0977

.0163

1.0506

1.1453

.2678

-.1418

ʌ

SEi(b)

.0097

.0001

.0094

.0100

.0219

-.1046

ʌ

SE2(b)

.0098

.0001

.0095

.0103

.2678

.1635

ʌ

SE3(b)

.0109

.0006

.0093

.0131

.2703

.2165

Note: SE1 is based on the usual covariance matrix formulae for the MG, FE and
POLS estimators.
SE2 is based on (10) and SE3 is based on (13).

21



More intriguing information

1. Wounds and reinscriptions: schools, sexualities and performative subjects
2. Placentophagia in Nonpregnant Nulliparous Mice: A Genetic Investigation1
3. The name is absent
4. Migration and employment status during the turbulent nineties in Sweden
5. The name is absent
6. The value-added of primary schools: what is it really measuring?
7. The name is absent
8. LOCAL CONTROL AND IMPROVEMENT OF COMMUNITY SERVICE
9. The name is absent
10. Recognizability of Individual Creative Style Within and Across Domains: Preliminary Studies