Rank-based estimating equations with general weight for accelerated failure time models: An induced smoothing approach

S. Chiou, Sangwook Kang, J. Yan

Research output: Contribution to journalArticle

10 Citations (Scopus)

Abstract

The induced smoothing technique overcomes the difficulties caused by the non-smoothness in rank-based estimating functions for accelerated failure time models, but it is only natural when the estimating function has Gehan's weight. For a general weight, the induced smoothing method does not provide smooth estimating functions that can be easily evaluated. We propose an iterative-induced smoothing procedure for general weights with the estimator from Gehan's weight initial value. The resulting estimators have the same asymptotic properties as those from the non-smooth estimating equations with the same weight. Their variances are estimated with an efficient resampling approach that avoids solving estimating equations repeatedly. The methodology is generalized to incorporate an additional weight to accommodate missing data and various sampling schemes. In a numerical study, the proposed estimators were obtained much faster without losing accuracy in comparison to those from non-smooth estimating equations, and the variance estimators provided good approximation of the variation in estimation. The methodology was applied to two real datasets, the first one from an adolescent depression study and the second one from a cancer study with missing covariates by design. The implementation is available in an R package aftgee.

Original languageEnglish
Pages (from-to)1495-1510
Number of pages16
JournalStatistics in Medicine
Volume34
Issue number9
DOIs
Publication statusPublished - 2015 Apr 30

Fingerprint

Accelerated Failure Time Model
Estimating Equation
Smoothing
Weights and Measures
Estimating Function
Nonsmooth Equations
Estimator
Missing Covariates
Smoothing Techniques
Smoothing Methods
Methodology
Variance Estimator
Resampling
Missing Data
Smooth function
Asymptotic Properties
Numerical Study
Cancer
Approximation
Neoplasms

All Science Journal Classification (ASJC) codes

  • Epidemiology
  • Statistics and Probability

Cite this

@article{aa4f3b8a88164ba9abc685e59c8131ba,
title = "Rank-based estimating equations with general weight for accelerated failure time models: An induced smoothing approach",
abstract = "The induced smoothing technique overcomes the difficulties caused by the non-smoothness in rank-based estimating functions for accelerated failure time models, but it is only natural when the estimating function has Gehan's weight. For a general weight, the induced smoothing method does not provide smooth estimating functions that can be easily evaluated. We propose an iterative-induced smoothing procedure for general weights with the estimator from Gehan's weight initial value. The resulting estimators have the same asymptotic properties as those from the non-smooth estimating equations with the same weight. Their variances are estimated with an efficient resampling approach that avoids solving estimating equations repeatedly. The methodology is generalized to incorporate an additional weight to accommodate missing data and various sampling schemes. In a numerical study, the proposed estimators were obtained much faster without losing accuracy in comparison to those from non-smooth estimating equations, and the variance estimators provided good approximation of the variation in estimation. The methodology was applied to two real datasets, the first one from an adolescent depression study and the second one from a cancer study with missing covariates by design. The implementation is available in an R package aftgee.",
author = "S. Chiou and Sangwook Kang and J. Yan",
year = "2015",
month = "4",
day = "30",
doi = "10.1002/sim.6415",
language = "English",
volume = "34",
pages = "1495--1510",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",
number = "9",

}

Rank-based estimating equations with general weight for accelerated failure time models : An induced smoothing approach. / Chiou, S.; Kang, Sangwook; Yan, J.

In: Statistics in Medicine, Vol. 34, No. 9, 30.04.2015, p. 1495-1510.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Rank-based estimating equations with general weight for accelerated failure time models

T2 - An induced smoothing approach

AU - Chiou, S.

AU - Kang, Sangwook

AU - Yan, J.

PY - 2015/4/30

Y1 - 2015/4/30

N2 - The induced smoothing technique overcomes the difficulties caused by the non-smoothness in rank-based estimating functions for accelerated failure time models, but it is only natural when the estimating function has Gehan's weight. For a general weight, the induced smoothing method does not provide smooth estimating functions that can be easily evaluated. We propose an iterative-induced smoothing procedure for general weights with the estimator from Gehan's weight initial value. The resulting estimators have the same asymptotic properties as those from the non-smooth estimating equations with the same weight. Their variances are estimated with an efficient resampling approach that avoids solving estimating equations repeatedly. The methodology is generalized to incorporate an additional weight to accommodate missing data and various sampling schemes. In a numerical study, the proposed estimators were obtained much faster without losing accuracy in comparison to those from non-smooth estimating equations, and the variance estimators provided good approximation of the variation in estimation. The methodology was applied to two real datasets, the first one from an adolescent depression study and the second one from a cancer study with missing covariates by design. The implementation is available in an R package aftgee.

AB - The induced smoothing technique overcomes the difficulties caused by the non-smoothness in rank-based estimating functions for accelerated failure time models, but it is only natural when the estimating function has Gehan's weight. For a general weight, the induced smoothing method does not provide smooth estimating functions that can be easily evaluated. We propose an iterative-induced smoothing procedure for general weights with the estimator from Gehan's weight initial value. The resulting estimators have the same asymptotic properties as those from the non-smooth estimating equations with the same weight. Their variances are estimated with an efficient resampling approach that avoids solving estimating equations repeatedly. The methodology is generalized to incorporate an additional weight to accommodate missing data and various sampling schemes. In a numerical study, the proposed estimators were obtained much faster without losing accuracy in comparison to those from non-smooth estimating equations, and the variance estimators provided good approximation of the variation in estimation. The methodology was applied to two real datasets, the first one from an adolescent depression study and the second one from a cancer study with missing covariates by design. The implementation is available in an R package aftgee.

UR - http://www.scopus.com/inward/record.url?scp=84926422216&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84926422216&partnerID=8YFLogxK

U2 - 10.1002/sim.6415

DO - 10.1002/sim.6415

M3 - Article

C2 - 25640630

AN - SCOPUS:84926422216

VL - 34

SP - 1495

EP - 1510

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 9

ER -