On testing the missing at random assumption

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

14 Citations (Scopus)

Abstract

Most approaches to learning from incomplete data are based on the assumption
that unobserved values are missing at random (mar). While the mar assumption, as such, is not testable, it can become testable in the context of other distributional assumptions, e.g. the naive Bayes assumption. In this paper we investigate a method for testing the mar assumption in the presence of other distributional constraints. We present methods to (approximately) compute a test statistic consisting of the ratio of two profile likelihood functions. This requires the optimization of the likelihood under no assumptionson the missingness mechanism, for which we use our recently proposed
AI \& M algorithm. We present experimental results on synthetic data that show that our approximate test statistic is a good indicator for whether data is mar relative to the given distributional assumptions.
Original languageEnglish
Title of host publicationMachine Learning: ECML 2006 : 17th European Conference on Machine Learning. Berlin, Germany, September 2006. Proceedings
Number of pages8
Publication date2006
Pages671-678
Publication statusPublished - 2006
EventEuropean Conference on Machine Learning - Berlin, Germany
Duration: 18 Sep 200622 Sep 2006

Conference

ConferenceEuropean Conference on Machine Learning
CountryGermany
CityBerlin
Period18/09/200622/09/2006

    Fingerprint

Cite this

Jaeger, M. (2006). On testing the missing at random assumption. In Machine Learning: ECML 2006: 17th European Conference on Machine Learning. Berlin, Germany, September 2006. Proceedings (pp. 671-678)