Over-Identification Tests in Earnings Functions with Fixed Effects.
The goodness-of-fit statistic for fixed effects models is shown to simplify to the degrees of freedom times the R(superscript "2") from a regression of analysis of covariance residuals on right hand side variables. This result is applied to test models for union wage effects. Identification and estimation of the return to schooling in models with fixed effects is also discussed. Analysis of covariance estimates of the returns to schooling are precisely estimated and roughly twice the size of ordinary least squares estimates. The goodness-of-fit test suggests that the fixed effects assumption is not appropriate for human capital earnings functions.
Year of publication: |
1991
|
---|---|
Authors: | Angrist, Joshua D ; Newey, Whitney K |
Published in: |
Journal of Business & Economic Statistics. - American Statistical Association. - Vol. 9.1991, 3, p. 317-23
|
Publisher: |
American Statistical Association |
Saved in:
Saved in favorites
Similar items by person
-
Semiparametric Efficiency Bounds.
Newey, Whitney K, (1990)
-
Semiparametric Estimation of Selection Models: Some Empirical Results.
Newey, Whitney K, (1990)
-
Generalized Method of Moments, Efficient Bootstrapping, and Improved Inference.
Brown, Bryan W, (2002)
- More ...