Optimal Model Selection in RDD and Related Settings Using Placebo Zones

73 Pages Posted: 31 Aug 2020

See all articles by Nathan Kettlewell

Nathan Kettlewell

University of Sydney

Peter Siminski

University of New South Wales (UNSW) - Social Policy Research Centre (SPRC)

Multiple version iconThere are 2 versions of this paper

Abstract

We propose a new model-selection algorithm for Regression Discontinuity Design, Regression Kink Design, and related IV estimators. Candidate models are assessed within a 'placebo zone' of the running variable, where the true effects are known to be zero. The approach yields an optimal combination of bandwidth, polynomial, and any other choice parameters. It can also inform choices between classes of models (e.g. RDD versus cohort-IV) and any other choices, such as covariates, kernel, or other weights. We use the approach to evaluate changes in Minimum Supervised Driving Hours in the Australian state of New South Wales. We also re-evaluate evidence on the effects of Head Start and Minimum Legal Drinking Age. We conclude with practical advice for researchers, including implications of treatment effect heterogeneity.

Keywords: regression discontinuity, regression kink, graduated driver licensing

JEL Classification: C13, C52, I18

Suggested Citation

Kettlewell, Nathan and Siminski, Peter, Optimal Model Selection in RDD and Related Settings Using Placebo Zones. IZA Discussion Paper No. 13639, Available at SSRN: https://ssrn.com/abstract=3682953

Nathan Kettlewell (Contact Author)

University of Sydney ( email )

Rm 370 Merewether (H04)
The University of Sydney
Sydney, NSW 2006 2008
Australia

HOME PAGE: http://sydney.edu.au/arts/economics/staff/profiles/nathan.kettlewell.php

Peter Siminski

University of New South Wales (UNSW) - Social Policy Research Centre (SPRC) ( email )

Sydney, NSW 2052
Australia
+61 2 9385 7827 (Phone)

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
1
Abstract Views
45
PlumX Metrics