TY - JOUR T1 - Itemwise conditionally independent nonresponse modeling for incomplete multivariate data JF - Biometrika Y1 - 2017 A1 - M. Sadinle A1 - J.P. Reiter KW - Loglinear model KW - Missing not at random KW - Missingness mechanism KW - Nonignorable KW - Nonparametric saturated KW - Sensitivity analysis AB - We introduce a nonresponse mechanism for multivariate missing data in which each study variable and its nonresponse indicator are conditionally independent given the remaining variables and their nonresponse indicators. This is a nonignorable missingness mechanism, in that nonresponse for any item can depend on values of other items that are themselves missing. We show that, under this itemwise conditionally independent nonresponse assumption, one can define and identify nonparametric saturated classes of joint multivariate models for the study variables and their missingness indicators. We also show how to perform sensitivity analysis to violations of the conditional independence assumptions encoded by this missingness mechanism. Throughout, we illustrate the use of this modeling approach with data analyses. VL - 104 UR - https://doi.org/10.1093/biomet/asw063 IS - 1 ER - TY - JOUR T1 - A nonparametric, multiple imputation-based method for the retrospective integration of data sets JF - Multivariate Behavioral Research Y1 - 2015 A1 - M.M. Carrig A1 - D. Manrique-Vallier A1 - K. Ranby A1 - J.P. Reiter A1 - R. Hoyle VL - 50 UR - http://www.tandfonline.com/doi/full/10.1080/00273171.2015.1022641 IS - 4 ER - TY - JOUR T1 - Semi-parametric selection models for potentially non-ignorable attrition in panel studies with refreshment samples JF - Political Analysis Y1 - 2015 A1 - Y. Si A1 - J.P. Reiter A1 - D.S. Hillygus VL - 23 UR - http://pan.oxfordjournals.org/cgi/reprint/mpu009?%20ijkey=joX8eSl6gyIlQKP&keytype=ref ER - TY - JOUR T1 - Stop or continue data collection: A nonignorable missing data approach for continuous variables JF - ArXiv Y1 - 2015 A1 - T. Paiva A1 - J.P. Reiter KW - Methodology AB - We present an approach to inform decisions about nonresponse followup sampling. The basic idea is (i) to create completed samples by imputing nonrespondents' data under various assumptions about the nonresponse mechanisms, (ii) take hypothetical samples of varying sizes from the completed samples, and (iii) compute and compare measures of accuracy and cost for different proposed sample sizes. As part of the methodology, we present a new approach for generating imputations for multivariate continuous data with nonignorable unit nonresponse. We fit mixtures of multivariate normal distributions to the respondents' data, and adjust the probabilities of the mixture components to generate nonrespondents' distributions with desired features. We illustrate the approaches using data from the 2007 U. S. Census of Manufactures. UR - http://arxiv.org/abs/1511.02189 IS - 1511.02189 ER - TY - JOUR T1 - Bayesian multiple imputation for large-scale categorical data with structural zeros JF - Survey Methodology Y1 - 2014 A1 - D. Manrique-Vallier A1 - J.P. Reiter VL - 40 UR - http://www.stat.duke.edu/ jerry/Papers/SurvMeth14.pdf ER - TY - CHAP T1 - Disclosure risk evaluation for fully synthetic data T2 - Privacy in Statistical Databases Y1 - 2014 A1 - J. Hu A1 - J.P. Reiter A1 - Q. Wang JF - Privacy in Statistical Databases PB - Springer CY - Heidelberg VL - 8744 ER - TY - JOUR T1 - Imputation of confidential data sets with spatial locations using disease mapping models JF - Statistics in Medicine Y1 - 2014 A1 - T. Paiva A1 - A. Chakraborty A1 - J.P. Reiter A1 - A.E. Gelfand VL - 33 ER -