Statistics and Its Interface

Volume 11 (2018)

Number 1

Regression analysis of incomplete data from event history studies with the proportional rates model

Pages: 91 – 97

DOI: http://dx.doi.org/10.4310/SII.2018.v11.n1.a8

Authors

Guanglei Yu (Department of Statistics, University of Missouri, Columbia, Mo., U.S.A.)

Liang Zhu (Biostatistics and Epidemiology Research Design, Health Science Center at Houston, University of Texas, Houston, Tx., U.S.A.)

Jianguo Sun (Department of Statistics, University of Missouri, Columbia, Mo., U.S.A.)

Leslie L. Robison (Department of Epidemiology and Cancer Control, St. Jude Children’s Research Hospital, Memphis, Tennessee, U.S.A.)

Abstract

Event history studies occur in many fields including epidemiology, sociology, and medical studies. They focus on the occurrences of some events of interest on subjects over time. One special type of data arising from such studies is incomplete mixed data, which is the mixed recurrent event data and panel count data. To deal with such type of data, we propose a proportional rates model and present a multiple imputation-based estimation procedure. One advantage of the proposed marginal model approach is that it can be easily implemented. To assess the performance of the procedure, a simulation study is conducted and indicates that it performs well for practical situations and can be more efficient than the existing method. The methodology is applied to a set of mixed data from a longitudinal cohort study.

Keywords

incomplete data, marginal model, multiple imputation, proportional rates model

Full Text (PDF format)

This work was partly supported by NIH Grant (R03 CA169150; R21 CA198641) to Zhu; and the funding from ALSAC and Cancer Center Support.

Paper received on 16 September 2016.