An improved survival estimator for censored medical costs with a kernel approach

Shuai Chen, Wenbin Lu, Hongwei Zhao

Research output: Contribution to journalArticle

Abstract

Cost assessment serves as an essential part in economic evaluation of medical interventions. In many studies, costs as well as survival data are frequently censored. Standard survival analysis techniques are often invalid for censored costs, due to the induced dependent censoring problem. Owing to high skewness in many cost data, it is desirable to estimate the median costs, which will be available with estimated survival function of costs. We propose a kernel-based survival estimator for costs, which is monotone, consistent, and more efficient than several existing estimators. We conduct numerical studies to examine the finite-sample performance of the proposed estimator.

Original languageEnglish (US)
Pages (from-to)1-15
Number of pages15
JournalCommunications in Statistics - Theory and Methods
DOIs
StateAccepted/In press - Nov 23 2017

Fingerprint

kernel
Estimator
Costs
Dependent Censoring
Survival Function
Survival Analysis
Survival Data
Skewness
Numerical Study
Monotone
Economics
Evaluation
Estimate

Keywords

  • Censored data
  • Kernel method
  • Median cost
  • Survival analysis

ASJC Scopus subject areas

  • Statistics and Probability

Cite this

An improved survival estimator for censored medical costs with a kernel approach. / Chen, Shuai; Lu, Wenbin; Zhao, Hongwei.

In: Communications in Statistics - Theory and Methods, 23.11.2017, p. 1-15.

Research output: Contribution to journalArticle

@article{2483daa863684497b08eded86846da7f,
title = "An improved survival estimator for censored medical costs with a kernel approach",
abstract = "Cost assessment serves as an essential part in economic evaluation of medical interventions. In many studies, costs as well as survival data are frequently censored. Standard survival analysis techniques are often invalid for censored costs, due to the induced dependent censoring problem. Owing to high skewness in many cost data, it is desirable to estimate the median costs, which will be available with estimated survival function of costs. We propose a kernel-based survival estimator for costs, which is monotone, consistent, and more efficient than several existing estimators. We conduct numerical studies to examine the finite-sample performance of the proposed estimator.",
keywords = "Censored data, Kernel method, Median cost, Survival analysis",
author = "Shuai Chen and Wenbin Lu and Hongwei Zhao",
year = "2017",
month = "11",
day = "23",
doi = "10.1080/03610926.2017.1400059",
language = "English (US)",
pages = "1--15",
journal = "Communications in Statistics - Theory and Methods",
issn = "0361-0926",
publisher = "Taylor and Francis Ltd.",

}

TY - JOUR

T1 - An improved survival estimator for censored medical costs with a kernel approach

AU - Chen, Shuai

AU - Lu, Wenbin

AU - Zhao, Hongwei

PY - 2017/11/23

Y1 - 2017/11/23

N2 - Cost assessment serves as an essential part in economic evaluation of medical interventions. In many studies, costs as well as survival data are frequently censored. Standard survival analysis techniques are often invalid for censored costs, due to the induced dependent censoring problem. Owing to high skewness in many cost data, it is desirable to estimate the median costs, which will be available with estimated survival function of costs. We propose a kernel-based survival estimator for costs, which is monotone, consistent, and more efficient than several existing estimators. We conduct numerical studies to examine the finite-sample performance of the proposed estimator.

AB - Cost assessment serves as an essential part in economic evaluation of medical interventions. In many studies, costs as well as survival data are frequently censored. Standard survival analysis techniques are often invalid for censored costs, due to the induced dependent censoring problem. Owing to high skewness in many cost data, it is desirable to estimate the median costs, which will be available with estimated survival function of costs. We propose a kernel-based survival estimator for costs, which is monotone, consistent, and more efficient than several existing estimators. We conduct numerical studies to examine the finite-sample performance of the proposed estimator.

KW - Censored data

KW - Kernel method

KW - Median cost

KW - Survival analysis

UR - http://www.scopus.com/inward/record.url?scp=85035081816&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85035081816&partnerID=8YFLogxK

U2 - 10.1080/03610926.2017.1400059

DO - 10.1080/03610926.2017.1400059

M3 - Article

AN - SCOPUS:85035081816

SP - 1

EP - 15

JO - Communications in Statistics - Theory and Methods

JF - Communications in Statistics - Theory and Methods

SN - 0361-0926

ER -