Markov counting models for correlated binary responses

Biostatistics
Forrest W Crawford, Daniel Zelterman

Abstract

We propose a class of continuous-time Markov counting processes for analyzing correlated binary data and establish a correspondence between these models and sums of exchangeable Bernoulli random variables. Our approach generalizes many previous models for correlated outcomes, admits easily interpretable parameterizations, allows different cluster sizes, and incorporates ascertainment bias in a natural way. We demonstrate several new models for dependent outcomes and provide algorithms for computing maximum likelihood estimates. We show how to incorporate cluster-specific covariates in a regression setting and demonstrate improved fits to well-known datasets from familial disease epidemiology and developmental toxicology.

References

Aug 1, 1992·Fundamental and Applied Toxicology : Official Journal of the Society of Toxicology·J F HolsonJ F Young
Sep 17, 2002·Biometrics·Chang Yu, Daniel Zelterman
May 15, 2003·American Journal of Epidemiology·Louise-Anne McNuttJean Paul Hafner
May 24, 2003·Biometrics·Catalina Stefanescu, Bruce W Turnbull
Jul 23, 2003·Statistics in Medicine·Jian-Lun Xu, Philip C Prorok
Mar 23, 2004·American Journal of Epidemiology·Guangyong Zou
Jul 12, 2008·Statistics in Medicine·Abigail G MatthewsRebecca A Betensky
Jan 1, 2008·Computational Statistics & Data Analysis·Chang Yu, Daniel Zelterman

❮ Previous
Next ❯

Citations

Jun 27, 2018·Wiley Interdisciplinary Reviews. Computational Statistics·Forrest W CrawfordMarc A Suchard

❮ Previous
Next ❯

Related Concepts

Related Feeds

Cancer Incidence & Mortality

Cancer has emerged as a global concern due to its increase in incidence and mortality. Efforts are underway to evaluate and develop action plans to reduce the global burden of cancer. Currently, lung cancer, breast cancer and prostate cancer are the leading causes of cancer mortality. Here is the latest research on cancer incidence and mortality.