Models for Categorical Data with Nonignorable Nonresponse

• Published in: Journal of the American Statistical Association Vol. 89; no. 425; pp. 44 - 52
• Main Authors: Park, Taesung, Brown, Morton B.
• Format: Journal Article
• Language: English
• Published: Alexandria, VA Taylor & Francis Group 01.03.1994; American Statistical Association; Taylor & Francis Ltd
• Subjects: Applications and Case Studies; Bayesian estimator; Children; Data models; Data smoothing; EM algorithm; Estimation bias; Estimators; Exact sciences and technology; Log-linear model; Mathematics; Maximum likelihood estimation; Maximum likelihood estimator; Missing data; Nonignorable missing data; Obesity; Parametric models; Prior distribution; Probability and statistics; Sampling theory, sample surveys; Sciences and techniques of general use; Smoothing; Statistical estimation; Statistics; Bayes estimation; Missing data; Categorical data; Smoothing; Prior distribution; Loglinear model; Maximum likelihood; Statistical simulation; Sample survey; EM algorithm; Mean square error

When categorical outcomes are subject to nonignorable nonresponse, log-linear models may be used to adjust for the nonresponse. The models are fitted to the data in an augmented frequency table in which one index corresponds to whether or not the subject is a respondent. The likelihood function is m...