Distinguishing “Missing at Random” and “Missing Completely at Random”
作者:
DanielF. Heitjan,
Srabashi Basu,
期刊:
The American Statistician
(Taylor Available online 1996)
卷期:
Volume 50,
issue 3
页码: 207-213
ISSN:0003-1305
年代: 1996
DOI:10.1080/00031305.1996.10474381
出版商: Taylor & Francis Group
关键词: Bayesian inference;Coarse data;Frequentist inference;Ignorability;Incomplete data;Likelihood inference;Missing data
数据来源: Taylor
摘要:
Missing at random (MAR) and missing completely at random (MCAR) are ignorability conditions—when they hold, they guarantee that certain kinds of inferences may be made without recourse to complicated missing-data modeling. In this article we review the definitions of MAR, MCAR, and their recent generalizations. We apply the definitions in three common incomplete-data examples, demonstrating by simulation the consequences of departures from ignorability. We argue that practitioners who face potentially non-ignorable incomplete data must consider both the mode of inference and the nature of the conditioning when deciding which ignorability condition to invoke.
点击下载:
PDF (765KB)
返 回