Regression Discontinuity in Time: Considerations for Empirical Applications

Abstract

Recent empirical work in several economic fields, particularly environmental and energy economics, has adapted the regression discontinuity (RD) framework to applications where time is the running variable and treatment begins at a particular threshold in time. In this guide for practitioners, we discuss several features of this regression discontinuity in time framework that differ from the more standard cross-sectional RD framework. First, many applications (particularly in environmental economics) lack cross-sectional variation and are estimated using observations far from the temporal threshold. This common empirical practice is hard to square with the assumptions of a cross-sectional RD, which is conceptualized for an estimation bandwidth shrinking even as the sample size increases. Second, estimates may be biased if the time-series properties of the data are ignored (for instance, in the presence of an autoregressive process), or more generally if short-run and long-run effects differ. Finally, tests for sorting or bunching near the threshold are often irrelevant, making the framework closer to an event study than a regression discontinuity design. Based on these features and motivated by hypothetical examples using air quality data, we offer suggestions for the empirical researcher wishing to use the RD in time framework.

The Trio Presidency and the Efficiency of Council Decision‐Making: An Empirical Study

“…We analyzed Hypothesis by transforming the data on the duration of individual proposals to a monthly average, classified by start date, and considered them as a time series. This format allowed us to apply an ITS design, similar to a regression discontinuity approach in time (for example, see Hausman and Rapson, ; Imbens and Lemieux, ). An ITS design is appropriate when, at a specific point in time, an intervention occurs, such as the enactment of a policy, treatment with a medicine or, in this case, a change in institutional rules as part of a change in a Treaty.…”

Who Benefits from Targeted Property Tax Relief? Evidence from Virginia Elections

“…For the more treated subsamples, we find little evidence of a shift in home composition coinciding with the 2010 election, with the only exception being the age of homes in high elderly areas. Nonetheless, we control for these characteristics to account for such differences within the hedonic framework, following the suggestion of Hausman and Rapson ().…”

“…We conduct a number of robustness checks that are conventional for studies that employ an RD design (Cattaneo, Titiunik, & Vazquez‐Bare, ; Hausman & Rapson, ), which we report in Appendix Figure and Appendix Table . Our results are robust to placebo discontinuities, alternative trend modeling, and varying bandwidths.…”

“…It is not uncommon to use time as a running variable, where the discontinuity is a point in time. See, for example, Fetter () and Hausman and Rapson () who discuss RD using a time running variable more generally.…”

Impact of Hurricane Maria on Beach Erosion in Puerto Rico: Remote Sensing and Causal Inference

“…A Regression discontinuity analysis operationalizes this intuition in a statistical framework and are commonly used in economics to evaluate the impact of policies characterized by sudden and arbitrary cutoffs (see Hausman & Rapson, ; Imbens & Lemieux, ; Lee & Lemieux, , for reviews and implementation guidelines). An important statistical challenge to address is the typically small sample of observations in the vicinity of the cutoffs.…”