Online Program Home
My Program

Abstract Details

Activity Number: 474 - Survey Sampling and Variance Estimation: Recent Innovations
Type: Contributed
Date/Time: Wednesday, July 31, 2019 : 8:30 AM to 10:20 AM
Sponsor: Survey Research Methods Section
Abstract #304685
Title: Expanding Variance Function Coverage in the Current Population Survey
Author(s): Justin McIllece*
Companies: U.S. Bureau of Labor Statistics
Keywords: CPS; GVF; variance functions

Generalized variance functions (GVFs) are used to produce official standard errors for the Current Population Survey (CPS) estimates included in The Employment Situation , a Principal Federal Economic Indicator (PFEI) published monthly by the Bureau of Labor Statistics. The parameters of the GVF models are also published and available to data users. However, variance function coverage beyond The Employment Situation is limited, and no GVF parameters are published for non-PFEI news releases. In this paper, recently developed models are extended in multiple dimensions, including: from time-dependent to time-robust parameters; from medians to other percentiles; from full-sample to partial-sample data; and from national to state-level estimates. Strengths and weaknesses of various research models are considered, specifically in relation to the quality of estimated standard errors and the convenience of the model form for parameter publication.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2019 program