BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Long Story Short: Omitted Variable Bias in Causal Machine Learning
  - Victor Chernozhukov (MIT)
DTSTART:20221007T150000Z
DTEND:20221007T160000Z
UID:TALK182714@talks.cam.ac.uk
CONTACT:Qingyuan Zhao
DESCRIPTION:We derive general\, yet simple\, sharp bounds on the size of t
 he omitted variable bias for a broad class of causal parameters that can b
 e identified as linear functionals of the conditional expectation function
  of the outcome. Such functionals encompass many of the traditional target
 s of investigation in causal inference studies\, such as\, for example\, (
 weighted) average of potential outcomes\, average treatment effects (inclu
 ding subgroup effects\, such as the effect on the treated)\, (weighted) av
 erage derivatives\, and policy effects from shifts in covariate distributi
 on -- all for general\, nonparametric causal models. Our construction reli
 es on the Riesz-Frechet representation of the target functional. Specifica
 lly\, we show how the bound on the bias depends only on the additional var
 iation that the latent variables create both in the outcome and in the Rie
 sz representer for the parameter of interest. Moreover\, in many important
  cases (e.g\, average treatment effects and avearage derivatives) the boun
 d is shown to depend on easily interpretable quantities that measure the e
 xplanatory power of the omitted variables. Therefore\, simple plausibility
  judgments on the maximum explanatory power of omitted variables (in expla
 ining treatment and outcome variation) are sufficient to place overall bou
 nds on the size of the bias. Furthermore\, we use debiased machine learnin
 g to provide flexible and efficient statistical inference on learnable com
 ponents of the bounds. Finally\, empirical examples demonstrate the useful
 ness of the approach.\n\nLink to paper: https://www.nber.org/papers/w30302
LOCATION:MR5\, Centre for Mathematical Sciences
END:VEVENT
END:VCALENDAR