Date of Award
Spring 2026
Document Type
Open Access Dissertation
Department
Statistics
First Advisor
Lianming Wang
Abstract
Current status data are frequently encountered in many real life cross-sectional epi- demiological and demographic studies. In such studies, each subject is examined only once, and the failure time of interest is never exactly observed but known to be either smaller or larger than the examination time based on the failure status. Cur- rent status data are a mixture of left-censored and right-censored observations for the failure times of all subjects with or without covariates. In some real life studies, the test procedure that is used to evaluate the failure status may be error-prone, and this leads to misclassifed failure status for some or all subjects. The resulting data are referred to as misclassifed current status data in the literature.
This dissertation introduces two novel expectation-maximization (EM) algorithms for the analysis of current status data with misclassifed responses, specifcally for the widely used semiparametric proportional odds model and the generalized odds ratio hazard model. One proposed method is developed using distinctive data aug- mentations with Gamma and Poisson latent variables, while the other employes Exponential and multinomial latent variables. Both methods use monotone splines to approximate the semiparametric components.
Extensive simulation studies demonstrate excellent estimation performance of the proposed methods, with comparisons conducted among the two proposed methods and an existing method. The proposed methods also exhibit advantages in computa- tional efciency, fexibility, straightforward implementation, and closed-form variance estimation. Extensions of the proposed methods to account for unknown testing ac- curacy are also considered for both methods. These techniques are also illustrated through applications to uterine fbroid data.
Rights
© 2026, Zhixin Chen
Recommended Citation
Chen, Z.(2026). Regression Analysis of Misclassified Current Status Data. (Doctoral dissertation). Retrieved from https://scholarcommons.sc.edu/etd/8859