A parametric approach to the estimation of cointegration vectors in panel data



Table 3: Sizes of the t-test

______Я =10______

______________.' = 20______________

2S-OLS

2S-HAC

DOLS

2S-OLS

2S-HAC

DOLS

Uncorrelated errors

20

0.098

0.097

0.157

0.140

0.115

0.162

30

0.057

0.052

0.124

0.057

0.046

0.115

50

0.044

0.041

0.081

0.039

0.036

0.078

100

0.058

0.064

0.045

0.072

0.069

0.045

Contemporaneously correlated errors

20

0.292

0.085

0.010

0.403

0.083

0.008

30

0.274

0.069

0.009

0.397

0.070

0.005

50

0.308

0.070

0.006

0.420

0.069

0.004

100

0.355

0.073

0.003

0.479

0.066

0.002

Note: Rejection frequency for tests of the null hypothesis b = 0 in model
(16). “2S-OLS” indicates rejection frequencies of a
t-test where the errors
are assumed to be i.i.d., whereas “2S-HAC” indicates a
t-test based on ro-
bust standard errors computed from (15). The column “DOLS” reports
rejection frequencies of the DOLS estimator assuming contemporaneously
uncorrelated errors. The nominal size is of the tests is 0.05 and 5000 repli-
cations are used to compute the rejection frequencies.

24



More intriguing information

1. Consciousness, cognition, and the hierarchy of context: extending the global neuronal workspace model
2. Portuguese Women in Science and Technology (S&T): Some Gender Features Behind MSc. and PhD. Achievement
3. Elicited bid functions in (a)symmetric first-price auctions
4. The name is absent
5. Smith and Rawls Share a Room
6. MATHEMATICS AS AN EXACT AND PRECISE LANGUAGE OF NATURE
7. PEER-REVIEWED FINAL EDITED VERSION OF ARTICLE PRIOR TO PUBLICATION
8. Long-Term Capital Movements
9. Wirtschaftslage und Reformprozesse in Estland, Lettland, und Litauen: Bericht 2001
10. Estimated Open Economy New Keynesian Phillips Curves for the G7