The Department of Economics Recruiting Seminar presents

"Jumpy or Kinky: Regression Discontinuity without the Discontinuity"
with Yingying Dong, Cal State Fullerton
 
Thursday, May 12, 2011
3:30-5:00 p.m.
Social Science Plaza B, Room 3266

Regression Discontinuity (RD) models identify local treatment effects by associating a discrete change in the mean outcome with a corresponding discrete change in the probability of treatment at a known threshold of a running variable. Dong shows that it is possible to identify the RD model treatment effect without a discontinuity. In particular, identification can come from a slope change (a kink) instead of a discrete level change (a jump) in the treatment probability. The intuition is based on L’hopital’s rule. The identification results can also be interpreted using instrumental variables models. Estimators are proposed that can be applied in the presence or absence of a discontinuity, by exploiting either a jump, or a kink, or both. The proposed estimators are applied to investigate the "retirement-consumption puzzle." In particular, Dong estimates the impact of retirement on household food consumption by exploiting changes in the retirement probability at 62, the early retirement age in the US.

For further information, please contact Gloria Simpson, simpsong@uci.edu.