Semiparametric Methods for Regression Under Two-Phase Sampling

• Activity Number: 340
• Type: Contributed
• Date/Time: Tuesday, July 31, 2012 : 10:30 AM to 12:20 PM
• Sponsor: Biometrics Section
• Author(s): Gustavo Amorim and Alastair John Scott and Christopher Wild*+
• Companies: University of Auckland and University of Auckland and University of Auckland
• Keywords: regression ; missing data ; estimating equations ; efficiency

Abstract

This paper extends the work in Jiang, Scott & Wild (2011), Scott & Wild (2011) on fitting regression models with response-biased two-phase samples, that is, samples where some or all the covariates are missing for some units and the probability that this happens depends in part on the value of the response of that unit. We look at a variety of methods based on estimating equations, at the relationship of these methods to semi-parametric efficient methods in cases where such methods exist, and show ways of obtaining efficiency gains that can sometimes be dramatic. The talk will concentrate on fitting linear models where the continuous response and some continuous covariates are available for all individuals but other covariates are observed only for subsets. Time permitting we will make connections with calibration and propensity scores.

Jiang Y., Scott A.J., Wild C.J. (2011). Adjusting for non-response in population-based case control studies. International Statistical Review, 79(2), 145-159. Scott A.J. and Wild C.J. (2011). Fitting regression models with response-biased samples. Canadian Journal of Statistics, 39, 519-536.