Search Results

Author: Huber, Martin
Resulting in 3 citations.
1. Farbmacher, Helmut
Huber, Martin
Laffers, Lukas
Langen, Henrika
Spindler, Martin
Causal Mediation Analysis with Double Machine Learning
Econometrics Journal published online (31 January 2022): DOI: 10.1093/ectj/utac003/6517682.
Also: https://academic.oup.com/ectj/advance-article/doi/10.1093/ectj/utac003/6517682
Cohort(s): NLSY97
Publisher: Royal Economic Society (RES)
Keyword(s): Health Care; Health/Health Status/SF-12 Scale; Insurance, Health; Statistical Analysis

Permission to reprint the abstract has not been received from the publisher.

This paper combines causal mediation analysis with double machine learning for a data-driven control of observed confounders in a high-dimensional setting. The average indirect effect of a binary treatment and the unmediated direct effect are estimated based on efficient score functions, which are robust w.r.t. misspecifications of the outcome, mediator, and treatment models. This property is key for selecting these models by double machine learning, which is combined with data splitting to prevent overfitting. We demonstrate that the effect estimators are asymptotically normal and n−1/2-consistent under specific regularity conditions and investigate the finite sample properties of the suggested methods in a simulation study when considering lasso as machine learner. We also provide an empirical application to the U.S. National Longitudinal Survey of Youth, assessing the indirect effect of health insurance coverage on general health operating via routine checkups as mediator, as well as the direct effect.
Bibliography Citation
Farbmacher, Helmut, Martin Huber, Lukas Laffers, Henrika Langen and Martin Spindler. "Causal Mediation Analysis with Double Machine Learning." Econometrics Journal published online (31 January 2022): DOI: 10.1093/ectj/utac003/6517682.
2. Huber, Martin
Causal Pitfalls in the Decomposition of Wage Gaps
Journal of Business and Economic Statistics 33,2 (2015): 179-191.
Also: http://www.tandfonline.com/doi/abs/10.1080/07350015.2014.937437
Cohort(s): NLSY79
Publisher: American Statistical Association
Keyword(s): Ethnic Differences; Racial Differences; Wage Gap

The decomposition of gender or ethnic wage gaps into explained and unexplained components (often with the aim to assess labor market discrimination) has been a major research agenda in empirical labor economics. This paper demonstrates that conventional decompositions, no matter whether linear or non-parametric, are equivalent to assuming a (probably too) simple model of mediation (aimed at assessing causal mechanisms) and may therefore lack causal interpretability. The reason is that decompositions typically control for post-birth variables that lie on the causal pathway from gender/ethnicity (which are determined at or even before birth) to wage but neglect potential endogeneity that may arise from this approach. Based on the newer literature on mediation analysis, we therefore provide more attractive identifying assumptions and discuss non-parametric identification based on reweighting.
Bibliography Citation
Huber, Martin. "Causal Pitfalls in the Decomposition of Wage Gaps." Journal of Business and Economic Statistics 33,2 (2015): 179-191.
3. Huber, Martin
Solovyeva, Anna
On the Sensitivity of Wage Gap Decompositions
Journal of Labor Research published online (7 May 2020): DOI: 10.1007/s12122-020-09302-7.
Also: https://link.springer.com/article/10.1007/s12122-020-09302-7
Cohort(s): NLSY79
Publisher: Springer
Keyword(s): Gender Differences; Statistical Analysis; Wage Gap

Permission to reprint the abstract has not been received from the publisher.

This paper investigates the sensitivity of average wage gap decompositions to methods resting on different assumptions regarding endogeneity of observed characteristics, sample selection into employment, and estimators' functional form. Applying five distinct decomposition techniques to estimate the gender wage gap in the U.S. using data from the National Longitudinal Survey of Youth 1979, we find that the magnitudes of the wage gap components are generally not stable across methods. Furthermore, the definition of the observed characteristics matters: merely including their current values (as frequently seen in wage decompositions) entails smaller explained and larger unexplained components than when including both their current values and histories in the analysis. Given the sensitivity of our results, we advise caution when using wage decompositions for policy recommendations.
Bibliography Citation
Huber, Martin and Anna Solovyeva. "On the Sensitivity of Wage Gap Decompositions." Journal of Labor Research published online (7 May 2020): DOI: 10.1007/s12122-020-09302-7.