Essays in causal inference

PDF
final_thesis.pdf - Published
Download (3MB)

URL:	https://madoc.bib.uni-mannheim.de/60097
URN:	urn:nbn:de:bsz:180-madoc-600973
Document Type:	Doctoral dissertation
Year of publication:	2021
Place of publication:	Mannheim
University:	Universität Mannheim
Evaluator:	Rothe, Christoph
Date of oral examination:	20 July 2021
Publication language:	English
Institution:	School of Law and Economics > Statistik (Rothe 2017-) Außerfakultäre Einrichtungen > Graduate School of Economic and Social Sciences - CDSE (Economics)
License:	Creative Commons Attribution 4.0 International (CC BY 4.0)
Subject:	330 Economics
Keywords (English):	causal inference
Abstract:	This thesis consists of three self-contained chapters: First, I develop a method to assess the sensitivity of local average treatment effect estimates to potential violations of the monotonicity assumption. Second, we propose new confidence sets for the parameter of interest in fuzzy regression discontinuity designs that are valid in a wide range of empirically relevant settings. Third, we propose a novel class of estimators that use pre-treatment covariates more efficiently than estimators that are commonly applied in the literature of regression discontinuity designs.