Joint modelling of recurrence and progression of adenomas: A latent variable approach

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

In this paper, we treat the number of recurrent adenomatous polyps as a latent variable and then use a mixture distribution to model the number of observed recurrent adenomatous polyps. This approach is equivalent to zero-inflated Poisson regression, which is a method used to analyse count data with excess zeros. In a zero-inflated Poisson model, a count response variable is assumed to be distributed as a mixture of a Poisson distribution and a distribution with point mass of one at zero. In many cancer studies, patients often have variable follow-up. When the disease of interest is subject to late onset, ignoring the length of follow-up will underestimate the recurrence rate. In this paper, we modify zero-inflated Poisson regression through a weight function to incorporate the length of follow-up into analysis. We motivate, develop, and illustrate the methods described here with an example from a colon cancer study.

Original languageEnglish (US)
Pages (from-to)201-215
Number of pages15
JournalStatistical Modelling
Volume5
Issue number3
DOIs
StatePublished - Oct 2005

Fingerprint

Joint Modeling
Latent Variables
Progression
Recurrence
Zero
Poisson Regression
Cancer
Mixture Distribution
Count Data
Poisson Model
Poisson distribution
Weight Function
Excess
Count
Latent variables
Modeling
Poisson regression

Keywords

  • Latent variable
  • Measurement error
  • Mixture distribution
  • Robust weight function
  • Variable follow-up
  • Zero-inflated poisson

ASJC Scopus subject areas

  • Statistics and Probability

Cite this

Joint modelling of recurrence and progression of adenomas : A latent variable approach. / Hsu, Chiu-Hsieh.

In: Statistical Modelling, Vol. 5, No. 3, 10.2005, p. 201-215.

Research output: Contribution to journalArticle

@article{c0e62bc1f39e412eac38ff3bde485f80,
title = "Joint modelling of recurrence and progression of adenomas: A latent variable approach",
abstract = "In this paper, we treat the number of recurrent adenomatous polyps as a latent variable and then use a mixture distribution to model the number of observed recurrent adenomatous polyps. This approach is equivalent to zero-inflated Poisson regression, which is a method used to analyse count data with excess zeros. In a zero-inflated Poisson model, a count response variable is assumed to be distributed as a mixture of a Poisson distribution and a distribution with point mass of one at zero. In many cancer studies, patients often have variable follow-up. When the disease of interest is subject to late onset, ignoring the length of follow-up will underestimate the recurrence rate. In this paper, we modify zero-inflated Poisson regression through a weight function to incorporate the length of follow-up into analysis. We motivate, develop, and illustrate the methods described here with an example from a colon cancer study.",
keywords = "Latent variable, Measurement error, Mixture distribution, Robust weight function, Variable follow-up, Zero-inflated poisson",
author = "Chiu-Hsieh Hsu",
year = "2005",
month = "10",
doi = "10.1191/1471082X05st094oa",
language = "English (US)",
volume = "5",
pages = "201--215",
journal = "Statistical Modelling",
issn = "1471-082X",
publisher = "SAGE Publications Ltd",
number = "3",

}

TY - JOUR

T1 - Joint modelling of recurrence and progression of adenomas

T2 - A latent variable approach

AU - Hsu, Chiu-Hsieh

PY - 2005/10

Y1 - 2005/10

N2 - In this paper, we treat the number of recurrent adenomatous polyps as a latent variable and then use a mixture distribution to model the number of observed recurrent adenomatous polyps. This approach is equivalent to zero-inflated Poisson regression, which is a method used to analyse count data with excess zeros. In a zero-inflated Poisson model, a count response variable is assumed to be distributed as a mixture of a Poisson distribution and a distribution with point mass of one at zero. In many cancer studies, patients often have variable follow-up. When the disease of interest is subject to late onset, ignoring the length of follow-up will underestimate the recurrence rate. In this paper, we modify zero-inflated Poisson regression through a weight function to incorporate the length of follow-up into analysis. We motivate, develop, and illustrate the methods described here with an example from a colon cancer study.

AB - In this paper, we treat the number of recurrent adenomatous polyps as a latent variable and then use a mixture distribution to model the number of observed recurrent adenomatous polyps. This approach is equivalent to zero-inflated Poisson regression, which is a method used to analyse count data with excess zeros. In a zero-inflated Poisson model, a count response variable is assumed to be distributed as a mixture of a Poisson distribution and a distribution with point mass of one at zero. In many cancer studies, patients often have variable follow-up. When the disease of interest is subject to late onset, ignoring the length of follow-up will underestimate the recurrence rate. In this paper, we modify zero-inflated Poisson regression through a weight function to incorporate the length of follow-up into analysis. We motivate, develop, and illustrate the methods described here with an example from a colon cancer study.

KW - Latent variable

KW - Measurement error

KW - Mixture distribution

KW - Robust weight function

KW - Variable follow-up

KW - Zero-inflated poisson

UR - http://www.scopus.com/inward/record.url?scp=26844540728&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=26844540728&partnerID=8YFLogxK

U2 - 10.1191/1471082X05st094oa

DO - 10.1191/1471082X05st094oa

M3 - Article

AN - SCOPUS:26844540728

VL - 5

SP - 201

EP - 215

JO - Statistical Modelling

JF - Statistical Modelling

SN - 1471-082X

IS - 3

ER -