:::

News & Events

:::

Speaker:Prof. Ching-Yun Wang (Division of Public Health Sciences, Fred Hutchinson Cancer Research Center)

Seminar
Poster:Webmasters事件日期:2019-03-29
 Topic:Multinomial Logistic Regression with Missing Outcome Data

Speaker:Prof. Ching-Yun Wang
(Division of Public Health Sciences, Fred Hutchinson Cancer Research Center)

Date Time:FRI. Mar 29, 2019, 10:40 AM - 12:00 PM 
 
Place: 4F-427, Assembly Building I

Abstract

Many diseases such as cancer and heart diseases are heterogeneous and it is of great interest to study the disease risk specific to the subtypes in relation to genetic and environmental risk factors. However, due to logistic and cost reasons, the subtype information for the disease is missing for some subjects. In this paper, we investigate methods for multinomial logistic regression with missing outcome data, including a bootstrap hot deck multiple imputation (BHMI), inverse probability weighted (IPW) and maximum likelihood (ML) estimators. These three methods are important approaches for missing data regression. The BHMI modifies the standard hot deck multiple imputation method such that it can provide valid confidence interval estimation. Under the situation when the covariates are discrete, the IPW and ML estimators are shown to be numerically identical. Extensive simulations show that all the three proposed methods yield unbiased estimators where the complete-case analysis can be biased if the missingness depends on the observed data. Further, all the proposed methods are more efficient than the complete-case analysis. The methods are applied to a colorectal cancer study.
 
Last modification time:2019-04-16 AM 11:34

  • recruiting animation-EN
  • NCTU
cron web_use_log