Analysis of Proportional Odds Models With Censoring and Errors-in-Covariates

Samiran Sinha, Yanyuan Ma

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

We propose a consistent method for estimating both the finite- and infinite-dimensional parameters of the proportional odds model when a covariate is subject to measurement error and time-to-events are subject to right censoring. The proposed method does not rely on the distributional assumption of the true covariate, which is not observed in the data. In addition, the proposed estimator does not require the measurement error to be normally distributed or to have any other specific distribution, and we do not attempt to assess the error distribution. Instead, we construct martingale-based estimators through inversion, using only the moment properties of the error distribution, estimable from multiple erroneous measurements of the true covariate. The theoretical properties of the estimators are established and the finite sample performance is demonstrated via simulations. We illustrate the usefulness of the method by analyzing a dataset from a clinical study on AIDS. Supplementary materials for this article are available online.

Original languageEnglish (US)
Pages (from-to)1301-1312
Number of pages12
JournalJournal of the American Statistical Association
Volume111
Issue number515
DOIs
StatePublished - Jul 2 2016

Fingerprint

Proportional Odds Model
Censoring
Covariates
Estimator
Measurement Error
Right Censoring
Martingale
Inversion
Moment
Simulation
Measurement error

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

@article{32836df2be034b3a8f73d9b14647625e,
title = "Analysis of Proportional Odds Models With Censoring and Errors-in-Covariates",
abstract = "We propose a consistent method for estimating both the finite- and infinite-dimensional parameters of the proportional odds model when a covariate is subject to measurement error and time-to-events are subject to right censoring. The proposed method does not rely on the distributional assumption of the true covariate, which is not observed in the data. In addition, the proposed estimator does not require the measurement error to be normally distributed or to have any other specific distribution, and we do not attempt to assess the error distribution. Instead, we construct martingale-based estimators through inversion, using only the moment properties of the error distribution, estimable from multiple erroneous measurements of the true covariate. The theoretical properties of the estimators are established and the finite sample performance is demonstrated via simulations. We illustrate the usefulness of the method by analyzing a dataset from a clinical study on AIDS. Supplementary materials for this article are available online.",
author = "Samiran Sinha and Yanyuan Ma",
year = "2016",
month = "7",
day = "2",
doi = "10.1080/01621459.2015.1093943",
language = "English (US)",
volume = "111",
pages = "1301--1312",
journal = "Journal of the American Statistical Association",
issn = "0162-1459",
publisher = "Taylor and Francis Ltd.",
number = "515",

}

Analysis of Proportional Odds Models With Censoring and Errors-in-Covariates. / Sinha, Samiran; Ma, Yanyuan.

In: Journal of the American Statistical Association, Vol. 111, No. 515, 02.07.2016, p. 1301-1312.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Analysis of Proportional Odds Models With Censoring and Errors-in-Covariates

AU - Sinha, Samiran

AU - Ma, Yanyuan

PY - 2016/7/2

Y1 - 2016/7/2

N2 - We propose a consistent method for estimating both the finite- and infinite-dimensional parameters of the proportional odds model when a covariate is subject to measurement error and time-to-events are subject to right censoring. The proposed method does not rely on the distributional assumption of the true covariate, which is not observed in the data. In addition, the proposed estimator does not require the measurement error to be normally distributed or to have any other specific distribution, and we do not attempt to assess the error distribution. Instead, we construct martingale-based estimators through inversion, using only the moment properties of the error distribution, estimable from multiple erroneous measurements of the true covariate. The theoretical properties of the estimators are established and the finite sample performance is demonstrated via simulations. We illustrate the usefulness of the method by analyzing a dataset from a clinical study on AIDS. Supplementary materials for this article are available online.

AB - We propose a consistent method for estimating both the finite- and infinite-dimensional parameters of the proportional odds model when a covariate is subject to measurement error and time-to-events are subject to right censoring. The proposed method does not rely on the distributional assumption of the true covariate, which is not observed in the data. In addition, the proposed estimator does not require the measurement error to be normally distributed or to have any other specific distribution, and we do not attempt to assess the error distribution. Instead, we construct martingale-based estimators through inversion, using only the moment properties of the error distribution, estimable from multiple erroneous measurements of the true covariate. The theoretical properties of the estimators are established and the finite sample performance is demonstrated via simulations. We illustrate the usefulness of the method by analyzing a dataset from a clinical study on AIDS. Supplementary materials for this article are available online.

UR - http://www.scopus.com/inward/record.url?scp=84991660197&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84991660197&partnerID=8YFLogxK

U2 - 10.1080/01621459.2015.1093943

DO - 10.1080/01621459.2015.1093943

M3 - Article

AN - SCOPUS:84991660197

VL - 111

SP - 1301

EP - 1312

JO - Journal of the American Statistical Association

JF - Journal of the American Statistical Association

SN - 0162-1459

IS - 515

ER -