Details
| Translated title of the contribution | Estimation of standard errors and treatment effects in empirical economics—methods and applications: Methoden und Anwendungen |
|---|---|
| Original language | Multiple languages |
| Pages (from-to) | 43-62 |
| Number of pages | 20 |
| Journal | Journal for Labour Market Research |
| Volume | 47 |
| Issue number | 1-2 |
| Publication status | Published - Mar 2014 |
Abstract
This paper discusses methodological problems of standard errors and treatment effects. First, heteroskedasticity- and cluster-robust estimates are considered as well as problems with Bernoulli distributed regressors, outliers and partially identified parameters. Second, procedures to determine treatment effects are analyzed. Four principles are in the focus: difference-in-differences estimators, matching procedures, treatment effects in quantile regression analysis and regression discontinuity approaches. These methods are applied to Cobb-Douglas functions using IAB establishment panel data. Different heteroskedasticity-consistent procedures lead to similar results of standard errors. Cluster-robust estimates show evident deviates. Dummies with a mean near 0.5 have a smaller variance of the coefficient estimates than others. Not all outliers have a strong influence on significance. New methods to handle the problem of partially identified parameters lead to more efficient estimates. The four discussed treatment procedures are applied to the question whether company-level pacts affect the output. In contrast to unconditional difference-in-differences and to estimates without matching the company-level effect is positive but insignificant if conditional difference-in-differences, nearest-neighbor or Mahalanobis metric matching is applied. The latter result has to be specified under quantile treatment effects analysis. The higher the quantile the higher is the positive company-level pact effect and there is a tendency from insignificant to significant effects. A sharp regression discontinuity analysis shows a structural break at a probability of 0.5 that a company-level pact exists. No specific effect of the Great Recession can be detected. Fuzzy regression discontinuity estimates reveal that the company-level pact effect is significantly lower in East than in West Germany.
ASJC Scopus subject areas
- Business, Management and Accounting(all)
- Industrial relations
- Economics, Econometrics and Finance(all)
- Economics and Econometrics
- Business, Management and Accounting(all)
- Organizational Behavior and Human Resource Management
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
In: Journal for Labour Market Research, Vol. 47, No. 1-2, 03.2014, p. 43-62.
Research output: Contribution to journal › Article › Research › peer review
}
TY - JOUR
T1 - Estimation of standard errors and treatment effects in empirical economics
T2 - methods and applications
AU - Hübler, Olaf
PY - 2014/3
Y1 - 2014/3
N2 - This paper discusses methodological problems of standard errors and treatment effects. First, heteroskedasticity- and cluster-robust estimates are considered as well as problems with Bernoulli distributed regressors, outliers and partially identified parameters. Second, procedures to determine treatment effects are analyzed. Four principles are in the focus: difference-in-differences estimators, matching procedures, treatment effects in quantile regression analysis and regression discontinuity approaches. These methods are applied to Cobb-Douglas functions using IAB establishment panel data. Different heteroskedasticity-consistent procedures lead to similar results of standard errors. Cluster-robust estimates show evident deviates. Dummies with a mean near 0.5 have a smaller variance of the coefficient estimates than others. Not all outliers have a strong influence on significance. New methods to handle the problem of partially identified parameters lead to more efficient estimates. The four discussed treatment procedures are applied to the question whether company-level pacts affect the output. In contrast to unconditional difference-in-differences and to estimates without matching the company-level effect is positive but insignificant if conditional difference-in-differences, nearest-neighbor or Mahalanobis metric matching is applied. The latter result has to be specified under quantile treatment effects analysis. The higher the quantile the higher is the positive company-level pact effect and there is a tendency from insignificant to significant effects. A sharp regression discontinuity analysis shows a structural break at a probability of 0.5 that a company-level pact exists. No specific effect of the Great Recession can be detected. Fuzzy regression discontinuity estimates reveal that the company-level pact effect is significantly lower in East than in West Germany.
AB - This paper discusses methodological problems of standard errors and treatment effects. First, heteroskedasticity- and cluster-robust estimates are considered as well as problems with Bernoulli distributed regressors, outliers and partially identified parameters. Second, procedures to determine treatment effects are analyzed. Four principles are in the focus: difference-in-differences estimators, matching procedures, treatment effects in quantile regression analysis and regression discontinuity approaches. These methods are applied to Cobb-Douglas functions using IAB establishment panel data. Different heteroskedasticity-consistent procedures lead to similar results of standard errors. Cluster-robust estimates show evident deviates. Dummies with a mean near 0.5 have a smaller variance of the coefficient estimates than others. Not all outliers have a strong influence on significance. New methods to handle the problem of partially identified parameters lead to more efficient estimates. The four discussed treatment procedures are applied to the question whether company-level pacts affect the output. In contrast to unconditional difference-in-differences and to estimates without matching the company-level effect is positive but insignificant if conditional difference-in-differences, nearest-neighbor or Mahalanobis metric matching is applied. The latter result has to be specified under quantile treatment effects analysis. The higher the quantile the higher is the positive company-level pact effect and there is a tendency from insignificant to significant effects. A sharp regression discontinuity analysis shows a structural break at a probability of 0.5 that a company-level pact exists. No specific effect of the Great Recession can be detected. Fuzzy regression discontinuity estimates reveal that the company-level pact effect is significantly lower in East than in West Germany.
KW - DiD estimators
KW - Matching
KW - Outliers
KW - Partially identified parameters
KW - Quantile regressions
KW - Regression discontinuity
KW - Standard errors
UR - http://www.scopus.com/inward/record.url?scp=84975217325&partnerID=8YFLogxK
U2 - 10.1007/s12651-013-0135-0
DO - 10.1007/s12651-013-0135-0
M3 - Article
AN - SCOPUS:84975217325
VL - 47
SP - 43
EP - 62
JO - Journal for Labour Market Research
JF - Journal for Labour Market Research
SN - 1614-3485
IS - 1-2
ER -