Many Average Partial Effects in l1-Regularized Binomial and Fractional Regressions
Harold D Chiang
Papers from arXiv.org
This paper studies inference for multiple/many average partial effects when outcome variable is binary or fractional under data rich, cluster sampling environments. The number of average partial effects of interest can be much larger than number of sample clusters. We propose a post-double-selection estimator as well as a Neyman orthogonal moment estimator, both based on $\ell_1$-penalization, and explore their asymptotic properties. The proposed estimators do not require oracle property for valid inference. We propose easy-to-implement algorithms for high-dimensional hypotheses testing and construction of simultaneously valid confidence intervals that are cluster-robust, robust against imperfect model selection using a new multiplier cluster bootstrap. We then apply algorithms to study multiple-testing problem of gendered language on the internet.
New Economics Papers: this item is included in nep-ecm
Date: 2018-12, Revised 2019-03
References: View references in EconPapers View complete reference list from CitEc
Citations: Track citations by RSS feed
Downloads: (external link)
http://arxiv.org/pdf/1812.09397 Latest version (application/pdf)
This item may be available elsewhere in EconPapers: Search for items with the same title.
Export reference: BibTeX
RIS (EndNote, ProCite, RefMan)
Persistent link: https://EconPapers.repec.org/RePEc:arx:papers:1812.09397
Access Statistics for this paper
More papers in Papers from arXiv.org
Bibliographic data for series maintained by arXiv administrators ().