BEGIN:VCALENDAR VERSION:2.0 PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4// BEGIN:VEVENT UID:20250630T100451EDT-2627nuPmLe@132.216.98.100 DTSTAMP:20250630T140451Z DESCRIPTION:\n Abstract:\n\n\nInformative selection\, in which the distribut ion of response variables given that they are sampled is different from th eir distribution in the population\, is pervasive in complex surveys. Fail ing to take such informativeness into account can produce severe inferenti al errors\, including biased and inconsistent estimation of population par ameters. While several parametric procedures exist to test for informative selection\, these methods are limited in scope and their parametric assum ptions are difficult to assess. We consider two classes of nonparametric t ests of informative selection. The first class is motivated by classic non parametric two-sample tests. We compare weighted and unweighted empirical distribution functions and obtain tests for informative selection that are analogous to Kolmogorov-Smirnov and Cramer-von Mises. For the second clas s of tests\, we adapt a kernel-based learning method that compares distrib utions based on their maximum mean discrepancy. The asymptotic distributio ns of the test statistics are established under the null hypothesis of non informative selection. Simulation results show that our tests have power c ompetitive with existing parametric tests in a correctly specified paramet ric setting\, and better than those tests under model misspecification. A recreational angling application illustrates the methodology.\n\n\n Speaker \n\n\nJay Breidt is a professor and associate chair in the Department of S tatistics. His research focuses on statistical inference in finite populat ion sampling and natural resources monitoring\; his specific emphasis is o n survey sampling\, time series and uncertainty quantification. Breidt is the associate editor for the Journal of Forecasting\, Environmental and Ec ological Statistics.\n\n\n Zoom Link\n\n Meeting ID: 939 8331 3215\n\n Passco de: 096952\n\n DTSTART:20210312T203000Z DTEND:20210312T213000Z SUMMARY:Jay Breidt (Colorado State University) URL:/mathstat/channels/event/jay-breidt-colorado-state -university-329400 END:VEVENT END:VCALENDAR