BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Drupal iCal API//EN
X-WR-CALNAME:Events items teaser
X-WR-TIMEZONE:America/Toronto
BEGIN:VTIMEZONE
TZID:America/Toronto
X-LIC-LOCATION:America/Toronto
BEGIN:DAYLIGHT
TZNAME:EDT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
DTSTART:20190310T070000
END:DAYLIGHT
BEGIN:STANDARD
TZNAME:EST
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
DTSTART:20191103T060000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
UID:68724c8dacf1a
DTSTART;TZID=America/Toronto:20200122T100000
SEQUENCE:0
TRANSP:TRANSPARENT
DTEND;TZID=America/Toronto:20200122T100000
URL:/statistics-and-actuarial-science/events/department
 -seminar-lin-liu-harvard-university
SUMMARY:Department seminar by Lin Liu\, Harvard University
CLASS:PUBLIC
DESCRIPTION:Summary \n\nTHE POSSIBILITY OF NEARLY ASSUMPTION-FREE INFERENCE
  IN CAUSAL\nINFERENCE\n\nIn causal effect estimation\, the state-of-the-ar
 t is the so-called\ndouble machine learning (DML) estimators\, which combi
 ne the benefit of\ndoubly robust estimation\, sample splitting and using m
 achine learning\nmethods to estimate nuisance parameters. The validity of 
 the\nconfidence interval associated with a DML estimator\, in most part\,\
 nrelies on the complexity of nuisance parameters and how close the\nmachin
 e learning estimators are to the nuisance parameters. Before we\nhave a co
 mplete understanding of the theory of many machine learning\nmethods incl
 uding deep neural networks\, even a DML estimator may\nhave a bias so l
 arge that prohibits valid inference. In this talk\,\nwe describe a nearly
  assumption-free procedure that can either\ncriticize the invalidity of t
 he Wald confidence interval associated\nwith the DML estimators of some 
 causal effect of interest or falsify\nthe certificates (i.e. the mathemat
 ical conditions) that\, if true\,\ncould ensure valid inference. Essentia
 lly\, we are testing the null\nhypothesis that if the bias of an estimator
  is smaller than a fraction\n$\\rho$ its standard error. Our test is valid
  under the null without\nrequiring any complexity (smoothness or sparsity)
  assumptions on the\nnuisance parameters or the properties of machine lear
 ning estimators\nand may have power to inform the analysts that they have 
 to do\nsomething else than DML estimators or Wald confidence intervals for
 \ninference purposes. This talk is based on joint work with Rajarshi\nMukh
 erjee and James M. Robins.\n 
DTSTAMP:20250712T115245Z
END:VEVENT
END:VCALENDAR