Survey Estimators That Respect Natural Orderings

• Activity Number: 615
• Type: Contributed
• Date/Time: Thursday, August 7, 2014 : 8:30 AM to 10:20 AM
• Sponsor: Survey Research Methods Section
• Author(s): Jiwen Wu*+ and Jean Opsomer and Mary Meyer
• Companies: and Colorado State University and Colorado State University
• Keywords: ordinal covariates ; constrained estimation ; isotonic regression ; domain estimation ; jackknife method

Abstract

Many variables in surveys contain natural orderings that should be respected in the estimates. For instance, the National Compensation Survey estimates mean wages for many job categories, and these mean wages are expected to be non-decreasing according to job level. In this type of situation, isotonic regression can be applied to give a constrained estimation satisfying the monotonicity. We combine domain estimation and the pooled adjacent algorithm to construct new design-weighted constrained estimators. Under some conditions on the sampling design and the population, the estimators are shown to be design consistent and asymptotically normal. The delete-d Jackknife method is suggested for variance estimation given that the constrained estimator is not necessarily smooth; simulation studies show that as long as d is chosen sufficiently large, the jackknife appears to estimate the variance well.