Transcription of Matias D. Cattaneo University of Michigan July 30, 2015
{{id}} {{{paragraph}}}
Example: bankruptcy
Matias D. Cattaneo University of Michigan July 30, 2015
Regression Discontinuity Designs in StataMatias D. CattaneoUniversity of MichiganJuly 30, 2015 OverviewMain goal: learn about treatment e ect of policy or treatment randomization available, easy to estimate treatment e treatment randomization not available, turn to observational on discontinuity (RD) and objective. Requires little information, if design be viewed as a local randomized to falsify, easy to : very local!Overview of RD package: estimation, inference and graphical presentation using localpolynomials, partitioning, and spacings : RD inference (point estimation and CI; classic, bias-corrected, robust).Irdbwselect: bandwidth or window selection (IK, CV, CCT).Irdplot: plots data (with optimal block length).rddensity package: discontinuity in density test at cuto ( manipulation testing)using novel local polynomial density : manipulation testing using local polynomial density : bandwidth or window package: covariate balance, binomial tests, randomization inferencemethods (window selection & inference).
I Instrumental variables. I Selection on observables. Regression discontinuity (RD) designs. ... Causal Inference: Y i(0) 6= Y ijT i= 0 and Y i(1) 6= Y ijT i= 1. ... F Pre-binned local polynomial method ΠMcCrary (2008). F New tuning-parameter-free method ΠCattaneo, Jansson and Ma ...
Tags:
Information
Domain:
Source:
Link to this page:
Please notify us if you found a problem with this document: