Conditions for Ignoring the Missing-Data Mechanism in Likelihood Inferences for Parameter Subsets

ABSTRACT For likelihood-based inferences from data with missing values, models are generally needed for both the data and the missing-data mechanism. However, modeling the mechanism can be challenging, and parameters are often poorly identified. Rubin in 1976 showed that for likelihood and Bayesian inference, sufficient conditions for ignoring the missing data mechanism are (a) the missing data are missing at random (MAR), in the sense that missingness does not depend on the missing values after conditioning on the observed data and (b) the parameters of the data model and the missingness mechanism are distinct, that is, there are no a priori ties, via parameter space restrictions or prior distributions, between these two sets of parameters. These conditions are sufficient but not always necessary, and they relate to the full vector of parameters of the data model. We propose definitions of partially MAR and ignorability for a subvector of the parameters of particular substantive interest, for direct likelihood/Bayesian and frequentist likelihood-based inference. We apply these definitions to a variety of examples. We also discuss conditioning on the pattern of missingness, as an alternative strategy for avoiding the need to model the missingness mechanism.

[1]  Roderick J. A. Little,et al.  Statistical Analysis with Missing Data: Little/Statistical Analysis with Missing Data , 2002 .

[2]  Prem K. Goel,et al.  Estimation of the Correlation Coefficient from a Broken Random Sample , 1980 .

[3]  D. Rubin.  Formalizing Subjective Notions about the Effect of Nonrespondents in Sample Surveys , 1977 .

[4]  Laura Ventura,et al.  Bayesian composite marginal likelihoods , 2011 .

[5]  D. Rubin.  INFERENCE AND MISSING DATA , 1975 .

[6]  Donald B. Rubin,et al.  Bayesian Inference for Causal Effects: The Role of Randomization , 1978 .

[7]  Shan Kang,et al.  Missing not at random models for masked clinical trials with dropouts , 2015, Clinical trials.

[8]  Donald B. Rubin,et al.  ‘Clarifying missing at random and related definitions, and implications when coupled with exchangeability’ , 2015 .

[9]  Dan Jackson,et al.  What Is Meant by "Missing at Random"? , 2013, 1306.2812.

[10]  R. Little.  Pattern-Mixture Models for Multivariate Incomplete Data , 1993 .

[11]  Ofer Harel,et al.  Partial and latent ignorability in missing-data problems , 2009 .

[12]  Roderick J. A. Little,et al.  Statistical Analysis with Missing Data , 1988 .

[13]  Laura Ventura,et al.  Prior Distributions From Pseudo-Likelihoods in the Presence of Nuisance Parameters , 2009 .

[14]  J. Cavanaugh,et al.  Partial Likelihood , 2018, Wiley StatsRef: Statistics Reference Online.

[15]  Joseph G. Ibrahim,et al.  A Bayesian justification of Cox's partial likelihood , 2003 .

[16]  Roderick J. A. Little,et al.  Subsample ignorable likelihood for regression analysis with missing data , 2011 .