2011
DOI: 10.1198/jasa.2011.tm10104
|View full text |Cite
|
Sign up to set email alerts
|

A Semiparametric Estimation of Mean Functionals With Nonignorable Missing Data

Abstract: Parameter estimation with non-ignorable missing data is a challenging problem in statistics. The fully parametric approach for joint modeling of the response model and the population model can produce results that are quite sensitive to the failure of the assumed model. We propose a more robust modeling approach by considering the model for the nonresponding part as an exponential tilting of the model for the responding part. The exponential tilting model can be justified under the assumption that the response… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

1
179
0

Year Published

2013
2013
2021
2021

Publication Types

Select...
7

Relationship

2
5

Authors

Journals

citations
Cited by 179 publications
(180 citation statements)
references
References 20 publications
1
179
0
Order By: Relevance
“…When MAR does not hold, method that brings in the exponential titling technique (Kim and Yu (2011)) can be developed accordingly. This will be investigated in the future.…”
Section: Discussionmentioning
confidence: 99%
“…When MAR does not hold, method that brings in the exponential titling technique (Kim and Yu (2011)) can be developed accordingly. This will be investigated in the future.…”
Section: Discussionmentioning
confidence: 99%
“…Also, as in Kim & Yu (2011), δ i is assumed to be independent of δ j for all i = j, and we let f j (y i |x i ) be the conditional distribution of Y i given x i and δ i = j, for j = 0, 1. Note that when f 0 (y i |x i ) = f 1 (y i |x i ), we recover the MAR assumption; that is, conditional on x i , δ i and y i are independent.…”
Section: Weighted Empirical Likelihood Rank Based Inferencementioning
confidence: 99%
“…Under this assumption, the probability that a response variable is missing depends on itself after controlling the predictors. As pointed out in Kim & Yu (2011), the MNAR condition exists, for example, in surveys of income, when the nonresponse rates tend to be higher for low socio-economic groups. This missingness type can be encountered in many fields of study such as social sciences, biomedical studies, agriculture economics and psychology among others.…”
Section: Introductionmentioning
confidence: 99%
See 2 more Smart Citations