External Validity in Fuzzy Regression Discontinuity Designs
By Marinho Bertanha, Guido W. Imbens. Journal of business & economic statistics, 2019
110 citations recorded by the source corpus.
Abstract Fuzzy regression discontinuity designs identify the local average treatment effect (LATE) for the subpopulation of compliers, and with forcing variable equal to the threshold. We develop methods that assess the external validity of LATE to other compliance groups at the threshold, and allow for identification away from the threshold. Specifically, we focus on the equality of outcome distributions between treated compliers and always-takers, and between untreated compliers and never-takers. These equalities imply continuity of expected outcomes conditional on both the forcing variable and the treatment status. We recommend that researchers plot these conditional expectations and test for discontinuities at the threshold to assess external validity. We provide new commands in STATA and MATLAB to implement our proposed procedures.
Subfields: Statistics and Probability
View Paper: https://www.semanticscholar.org/paper/1b74ab893965c0075e846aebaed7e0484781ba34
Citations in the corpus
- A Practical Introduction to Regression Discontinuity Designs (2019, unknown): “…2015; Wing and Cook, 2013), (ii) local independence assumptions (Angrist and Rokkanen, 2015) (iii) exploiting specific features of the design such as imperfect compliance (Bertanha and Imbens, 2019), or (iv) the presence of multiple cutoffs (Cattaneo, Keele, Titiunik, and Vazquez-Bare, 2016, 2019).”
- Impossible inference in econometrics: Theory and applications (2016, unknown): “…include estimation of derivatives of the treatment effect at the cutoff by Dong (2016) and Dong and Lewbel (2015); tests for homogeneity of treatment effects in fuzzy RDD by Bertanha and Imbens (2019); and estimation of average treatment effects in RDD with variation in cutoff values by Bertanha (2019).”
- An impact assessment of machine learning risk forecasts on parole board decisions and recidivism (2017, unknown): “One might think that this opens the door for a fuzzy regression discontinuity design (FRDD), but then one must assume that there is a discontinuity in the probability of having the forecasts available (Hahn et al. 2001; Bertanha and Imbens 2014).”
- Econometric Methods for Program Evaluation (2018, unknown): “, Europe, and elsewhere (Bothwell et al., 2016). While randomized assignment is often viewed as the basis of a high standard of scientific evidence, this view is by no means universal. Cartwright (2007), Heckman and Vytlacil (2007), and Deaton (2010), among others, challenge the notion that randomiz”