Inference in Regression Models with Missing Covariates

$384,746R01FY2009CANIH

Univ Of North Carolina Chapel Hill, Chapel Hill NC

Investigators

Linked publications & trials

Paper 34326706 Paper 29422418 Paper 27974868 Paper 27062562 Paper 27057264 Paper 26997848 Paper 26903666 Paper 26753050 Paper 26527844 Paper 26257438 Paper 26089629 Paper 25772374 Paper 25762198 Paper 25388125 Paper 25349462 Paper 25309677 Paper 25044061 Paper 25041037 Paper 24948838 Paper 24753718 Paper 24533040 Paper 24453379 Paper 24343859 Paper 24193892 Paper 24128143 Paper 23794769 Paper 23620446 Paper 23580436 Paper 23543121 Paper 23399928 Paper 23389549 Paper 23194570 Paper 23152646 Paper 23008683 Paper 22829358 Paper 22649133 Paper 22484596 Paper 22385010 Paper 22376150 Paper 21955084 Paper 21860598 Paper 21787385 Paper 21765885 Paper 21710558 Paper 21705484 Paper 21533000 Paper 21532998 Paper 21361889 Paper 21218187 Paper 21187451 Paper 21155748 Paper 21151838 Paper 21132093 Paper 20960582 Paper 20825395 Paper 20801907 Paper 20662831 Paper 20585409 Paper 20560937 Paper 20526807 Paper 20414804 Paper 20336190 Paper 20205269 Paper 20157003 Paper 20037674 Paper 19890478 Paper 19802375 Paper 19694273 Paper 19693282 Paper 19691015 Paper 19657473 Paper 19525084 Paper 19459831 Paper 19436775 Paper 19436774 Paper 19289959 Paper 19210738 Paper 19197247 Paper 19169388 Paper 19156361 Paper 19028825 Paper 19018265 Paper 18836829 Paper 18815102 Paper 18422991 Paper 17911498 Paper 17688501 Paper 17649909 Paper 17425636 Paper 17403100 Paper 17401683 Paper 17136621 Paper 16542244 Paper 16428260 Paper 16345018 Paper 16135028 Paper 16135024 Paper 16011696 Paper 15737075 Paper 14969495

Abstract

In this proposal, we propose Bayesian and frequentist methodology for local influence diagnostics and develop model assessment tools for complete data settings as well as in the presence of missing covariate and/or response data for a variety of statistical models, including generalized linear models, models for longitudinal data, and survival model. In Specific Aim 1, we develop frequentist local influence measures and goodness of fit statistics based on the general local influence development of Cook (1986), and discuss these measures for i) linear models with missing at random (MAR) and nonignorably missing covariates and ii) generalized linear models with MAR and nonignorably missing covariates. For Specific Aim 2, we develop new classes of Bayesian case influence diagnostics for the complete data setting then generalize these diagnostics to the missing data framework. The proposed methodologies in Aims 1-2 are primarily motivated from several studies in the PI's collaborative work.

View original record on NIH RePORTER →