Back to Search
Start Over
Bias Reduction in Logistic Regression with Missing Responses When the Missing Data Mechanism is Nonignorable
- Source :
- The American Statistician. 73:340-349
- Publication Year :
- 2018
- Publisher :
- Informa UK Limited, 2018.
-
Abstract
- In logistic regression with nonignorable missing responses, Ibrahim and Lipsitz (1996)) proposed a method for estimating regression parameters. It is known that the regression estimates obtained by using this method are biased when the sample size is small. Also, another complexity arises when the iterative estimation process encounters separation in estimating regression coefficients. In this article we propose a method to improve the estimation of regression coefficients. In our likelihood based method, we penalize the likelihood by multiplying it by a non-informative Jeffreys prior as a penalty term. The proposed method reduces bias and is able to handle the issue of separation. Simulation results show substantial bias reduction for the proposed method as compared to the existing method. Analyses using real world data also support the simulation findings. An R package called brlrmr is developed implementing the proposed method and the Ibrahim and Lipsitz method.
- Subjects :
- Statistics and Probability
Mechanism (biology)
General Mathematics
05 social sciences
Separation (statistics)
Logistic regression
Missing data
01 natural sciences
050105 experimental psychology
Bias reduction
Regression
Statistics::Computation
010104 statistics & probability
Expectation–maximization algorithm
Statistics
Statistics::Methodology
0501 psychology and cognitive sciences
0101 mathematics
Statistics, Probability and Uncertainty
Mathematics
Subjects
Details
- ISSN :
- 15372731 and 00031305
- Volume :
- 73
- Database :
- OpenAIRE
- Journal :
- The American Statistician
- Accession number :
- edsair.doi.dedup.....3bc7d7254d00c80e9a3861c2ff09abb0
- Full Text :
- https://doi.org/10.1080/00031305.2017.1407359