IASS Webinar 29: Fitting Classification Trees to Complex Survey Data

Map Unavailable

Date/Time
Date(s) - 31/05/2023
2:00 pm - 3:30 pm

Category(ies) No Categories


IASS Webinar 29: Fitting Classification Trees to Complex Survey Data

Date and Time: 31 May 2023  2pm – 3:30pm (CET)

Speaker: Jean Opsomer, Westat

 

All are invited to the webinar, organised by the International Association for Survey Statisticians.

Please register  for the IASS Webinar at:  https://bit.ly/IASS-webinar-29  

After registering, you will receive a confirmation email containing information about joining the webinar. There will be time for questions. The webinar will be recorded and made available on the IASS and ISI web site. See below for the abstract and biography of the speakers.

 

Webinar Abstract

Classification tree algorithms are a convenient method to perform variable selection and obtain interpretable structures relating covariates and an outcome of interest.  When fitting classification trees to survey data, it is common to ignore sampling weights as well other design characteristics such as stratification and clustering.  However, unless the survey design is uninformative, there is a risk that the inference for the classification tree is incorrect. A particular application in which this is a concern is the construction of nonresponse adjustment cells, a key step in the development of survey weights.  We propose an extension of the popular Chi-square Automatic Interaction Detector (CHAID) approach that accounts for the design by applying a Rao-Scott correction in its classification criterion. We discuss the statistical properties of the resulting algorithm under a design-based framework.  We compare its performance to existing weighted and unweighted algorithms, and we illustrate the use of the method using data from the U.S. American Community Survey.

 

Biography

Jean Opsomer is Vice President at Westat, where he is responsible for the statistical methodology of several large-scale survey projects, and Adjunct Professor in the Department of Mathematics at the University of Maryland, College Park. Previously, he was a faculty member in the Departments of Statistics at Colorado State University and Iowa State University.  His main research interests are in survey statistics and nonparametric statistics, as well as the intersection of the two.  He is a Fellow of the American Statistical Association and the Institute of Mathematical Statistics, and an Elected Member of the International Statistical Institute.