2017
DOI: 10.3386/w23602
|View full text |Cite
|
Sign up to set email alerts
|

Regression Discontinuity in Time: Considerations for Empirical Applications

Abstract: for helpful comments. We also thank UC Berkeley's Energy Institute at Haas, where Rapson was a visitor while this research was conducted. The views expressed herein are those of the authors and do not necessarily reflect the views of the National Bureau of Economic Research. NBER working papers are circulated for discussion and comment purposes. They have not been peer-reviewed or been subject to the review by the NBER Board of Directors that accompanies official NBER publications.

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

4
167
0

Year Published

2018
2018
2022
2022

Publication Types

Select...
8

Relationship

0
8

Authors

Journals

citations
Cited by 124 publications
(171 citation statements)
references
References 13 publications
4
167
0
Order By: Relevance
“…However, and contrary to event studies where a linear model is the most frequent specification, we use a flexible approach that incorporates unobserved time series 3 We should note that a reviewer suggested a regression discontinuity approach using time as assignment variable. Besides specific methodological problems, such as the non-randomized nature of the assignment variable and the specification of the time series model (Lee and Lemieux, 2010;Hausman and Rapson, 2018), this approach would produce an estimate of the average effect preventing any inference on the distribution of the impact.…”
Section: Measuring Informational Spilloversmentioning
confidence: 99%
“…However, and contrary to event studies where a linear model is the most frequent specification, we use a flexible approach that incorporates unobserved time series 3 We should note that a reviewer suggested a regression discontinuity approach using time as assignment variable. Besides specific methodological problems, such as the non-randomized nature of the assignment variable and the specification of the time series model (Lee and Lemieux, 2010;Hausman and Rapson, 2018), this approach would produce an estimate of the average effect preventing any inference on the distribution of the impact.…”
Section: Measuring Informational Spilloversmentioning
confidence: 99%
“…As suggested in Hausman and Rapson [2018], we account for seasonality in the outcomes of interests. In other word, the variable y bt corresponds to the residuals that come from regressing the outcome on month dummies and quadratic functions of temperature and precipitation as well as beat fixed-effects.…”
Section: Sample Selectionmentioning
confidence: 99%
“…This estimated difference is interpreted as the a point in time. See, for example, Fetter (2013) and Hausman and Rapson (2017) who discuss RD using a time running variable more generally. 10 See Imberman and Lovenheim (2015) for a related design that uses a difference-in-differences (DD) approach to estimate the effect of a sudden news release of school ratings on home prices.…”
Section: Baseline Rd Approachmentioning
confidence: 99%
“…We conduct a number of robustness checks that are conventional for studies that employ an RD design (Cattaneo, Titiunik, & Vazquez-Bare, 2017;Hausman & Rapson, 2017), which we report in Appendix Figure A8 and Appendix Table A9. Our results are robust to placebo discontinuities, alternative trend modeling, and varying bandwidths.…”
Section: Alternative Rd Specificationsmentioning
confidence: 99%