Semiparametric estimation for the additive hazards model with left-truncated and right-censored data

Chiung Yu Huang, Jing Qin

Research output: Contribution to journalArticle

Abstract

Survival data from prevalent cases collected under a cross-sectional sampling scheme are subject to left-truncation. When fitting an additive hazards model to left-truncated data, the conditional estimating equation method (Lin & Ying, 1994), obtained by modifying the risk sets to account for left-truncation, can be very inefficient, as the marginal likelihood of the truncation times is not used in the estimation procedure. In this paper, we use a pairwise pseudolikelihood to eliminate nuisance parameters from the marginal likelihood and, by combining the marginal pairwise pseudo-score function and the conditional estimating function, propose an efficient estimator for the additive hazards model. The proposed estimator is shown to be consistent and asymptotically normally distributed with a sandwich-type covariance matrix that can be consistently estimated. Simulation studies show that the proposed estimator is more efficient than its competitors. A data analysis illustrates application of the method.

Original languageEnglish (US)
Pages (from-to)877-888
Number of pages12
JournalBiometrika
Volume100
Issue number4
DOIs
StatePublished - Dec 2013

Fingerprint

Additive Hazards Model
Left Truncation
Marginal Likelihood
Semiparametric Estimation
Right-censored Data
Proportional Hazards Models
Pairwise
Hazards
Truncated Data
Estimator
Pseudo-likelihood
Efficient Estimator
Estimating Function
Score Function
Estimating Equation
Survival Data
Nuisance Parameter
Sandwich
sandwiches
Covariance matrix

Keywords

  • Canadian Study of Health and Aging
  • Composite likelihood
  • Estimating equation
  • Martingale
  • Prevalent sampling

ASJC Scopus subject areas

  • Agricultural and Biological Sciences(all)
  • Agricultural and Biological Sciences (miscellaneous)
  • Statistics and Probability
  • Mathematics(all)
  • Applied Mathematics
  • Statistics, Probability and Uncertainty

Cite this

Semiparametric estimation for the additive hazards model with left-truncated and right-censored data. / Huang, Chiung Yu; Qin, Jing.

In: Biometrika, Vol. 100, No. 4, 12.2013, p. 877-888.

Research output: Contribution to journalArticle

Huang, Chiung Yu ; Qin, Jing. / Semiparametric estimation for the additive hazards model with left-truncated and right-censored data. In: Biometrika. 2013 ; Vol. 100, No. 4. pp. 877-888.
@article{716d747da0564ba888fb63b20f22e9bc,
title = "Semiparametric estimation for the additive hazards model with left-truncated and right-censored data",
abstract = "Survival data from prevalent cases collected under a cross-sectional sampling scheme are subject to left-truncation. When fitting an additive hazards model to left-truncated data, the conditional estimating equation method (Lin & Ying, 1994), obtained by modifying the risk sets to account for left-truncation, can be very inefficient, as the marginal likelihood of the truncation times is not used in the estimation procedure. In this paper, we use a pairwise pseudolikelihood to eliminate nuisance parameters from the marginal likelihood and, by combining the marginal pairwise pseudo-score function and the conditional estimating function, propose an efficient estimator for the additive hazards model. The proposed estimator is shown to be consistent and asymptotically normally distributed with a sandwich-type covariance matrix that can be consistently estimated. Simulation studies show that the proposed estimator is more efficient than its competitors. A data analysis illustrates application of the method.",
keywords = "Canadian Study of Health and Aging, Composite likelihood, Estimating equation, Martingale, Prevalent sampling",
author = "Huang, {Chiung Yu} and Jing Qin",
year = "2013",
month = "12",
doi = "10.1093/biomet/ast039",
language = "English (US)",
volume = "100",
pages = "877--888",
journal = "Biometrika",
issn = "0006-3444",
publisher = "Oxford University Press",
number = "4",

}

TY - JOUR

T1 - Semiparametric estimation for the additive hazards model with left-truncated and right-censored data

AU - Huang, Chiung Yu

AU - Qin, Jing

PY - 2013/12

Y1 - 2013/12

N2 - Survival data from prevalent cases collected under a cross-sectional sampling scheme are subject to left-truncation. When fitting an additive hazards model to left-truncated data, the conditional estimating equation method (Lin & Ying, 1994), obtained by modifying the risk sets to account for left-truncation, can be very inefficient, as the marginal likelihood of the truncation times is not used in the estimation procedure. In this paper, we use a pairwise pseudolikelihood to eliminate nuisance parameters from the marginal likelihood and, by combining the marginal pairwise pseudo-score function and the conditional estimating function, propose an efficient estimator for the additive hazards model. The proposed estimator is shown to be consistent and asymptotically normally distributed with a sandwich-type covariance matrix that can be consistently estimated. Simulation studies show that the proposed estimator is more efficient than its competitors. A data analysis illustrates application of the method.

AB - Survival data from prevalent cases collected under a cross-sectional sampling scheme are subject to left-truncation. When fitting an additive hazards model to left-truncated data, the conditional estimating equation method (Lin & Ying, 1994), obtained by modifying the risk sets to account for left-truncation, can be very inefficient, as the marginal likelihood of the truncation times is not used in the estimation procedure. In this paper, we use a pairwise pseudolikelihood to eliminate nuisance parameters from the marginal likelihood and, by combining the marginal pairwise pseudo-score function and the conditional estimating function, propose an efficient estimator for the additive hazards model. The proposed estimator is shown to be consistent and asymptotically normally distributed with a sandwich-type covariance matrix that can be consistently estimated. Simulation studies show that the proposed estimator is more efficient than its competitors. A data analysis illustrates application of the method.

KW - Canadian Study of Health and Aging

KW - Composite likelihood

KW - Estimating equation

KW - Martingale

KW - Prevalent sampling

UR - http://www.scopus.com/inward/record.url?scp=84890320821&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84890320821&partnerID=8YFLogxK

U2 - 10.1093/biomet/ast039

DO - 10.1093/biomet/ast039

M3 - Article

C2 - 26246622

AN - SCOPUS:84890320821

VL - 100

SP - 877

EP - 888

JO - Biometrika

JF - Biometrika

SN - 0006-3444

IS - 4

ER -