Semiparametric single-index model for estimating optimal individualized treatment strategy

Rui Song, Shikai Luo, Donglin Zeng, Hao Zhang, Wenbin Lu, Zhiguo Li

Research output: Contribution to journalArticle

Abstract

Different from the standard treatment discovery framework which is used for finding single treatments for a homogenous group of patients, personalized medicine involves finding therapies that are tailored to each individual in a heterogeneous group. In this paper, we propose a new semiparametric additive single-index model for estimating individualized treatment strategy. The model assumes a flexible and nonparametric link function for the interaction between treatment and predictive covariates. We estimate the rule via monotone B-splines and establish the asymptotic properties of the estimators. Both simulations and an real data application demonstrate that the proposed method has a competitive performance.

Original languageEnglish (US)
Pages (from-to)364-384
Number of pages21
JournalElectronic Journal of Statistics
Volume11
Issue number1
DOIs
StatePublished - 2017

Fingerprint

Single-index Model
Semiparametric Model
Link Function
Additive Models
B-spline
Medicine
Asymptotic Properties
Therapy
Covariates
Monotone
Estimator
Strategy
Index model
Interaction
Estimate
Demonstrate
Simulation

Keywords

  • Personalized medicine
  • Semiparametric inference
  • Single index model

ASJC Scopus subject areas

  • Statistics and Probability

Cite this

Semiparametric single-index model for estimating optimal individualized treatment strategy. / Song, Rui; Luo, Shikai; Zeng, Donglin; Zhang, Hao; Lu, Wenbin; Li, Zhiguo.

In: Electronic Journal of Statistics, Vol. 11, No. 1, 2017, p. 364-384.

Research output: Contribution to journalArticle

Song, Rui ; Luo, Shikai ; Zeng, Donglin ; Zhang, Hao ; Lu, Wenbin ; Li, Zhiguo. / Semiparametric single-index model for estimating optimal individualized treatment strategy. In: Electronic Journal of Statistics. 2017 ; Vol. 11, No. 1. pp. 364-384.
@article{d489cddaa3a8409f978b4ba401b01652,
title = "Semiparametric single-index model for estimating optimal individualized treatment strategy",
abstract = "Different from the standard treatment discovery framework which is used for finding single treatments for a homogenous group of patients, personalized medicine involves finding therapies that are tailored to each individual in a heterogeneous group. In this paper, we propose a new semiparametric additive single-index model for estimating individualized treatment strategy. The model assumes a flexible and nonparametric link function for the interaction between treatment and predictive covariates. We estimate the rule via monotone B-splines and establish the asymptotic properties of the estimators. Both simulations and an real data application demonstrate that the proposed method has a competitive performance.",
keywords = "Personalized medicine, Semiparametric inference, Single index model",
author = "Rui Song and Shikai Luo and Donglin Zeng and Hao Zhang and Wenbin Lu and Zhiguo Li",
year = "2017",
doi = "10.1214/17-EJS1226",
language = "English (US)",
volume = "11",
pages = "364--384",
journal = "Electronic Journal of Statistics",
issn = "1935-7524",
publisher = "Institute of Mathematical Statistics",
number = "1",

}

TY - JOUR

T1 - Semiparametric single-index model for estimating optimal individualized treatment strategy

AU - Song, Rui

AU - Luo, Shikai

AU - Zeng, Donglin

AU - Zhang, Hao

AU - Lu, Wenbin

AU - Li, Zhiguo

PY - 2017

Y1 - 2017

N2 - Different from the standard treatment discovery framework which is used for finding single treatments for a homogenous group of patients, personalized medicine involves finding therapies that are tailored to each individual in a heterogeneous group. In this paper, we propose a new semiparametric additive single-index model for estimating individualized treatment strategy. The model assumes a flexible and nonparametric link function for the interaction between treatment and predictive covariates. We estimate the rule via monotone B-splines and establish the asymptotic properties of the estimators. Both simulations and an real data application demonstrate that the proposed method has a competitive performance.

AB - Different from the standard treatment discovery framework which is used for finding single treatments for a homogenous group of patients, personalized medicine involves finding therapies that are tailored to each individual in a heterogeneous group. In this paper, we propose a new semiparametric additive single-index model for estimating individualized treatment strategy. The model assumes a flexible and nonparametric link function for the interaction between treatment and predictive covariates. We estimate the rule via monotone B-splines and establish the asymptotic properties of the estimators. Both simulations and an real data application demonstrate that the proposed method has a competitive performance.

KW - Personalized medicine

KW - Semiparametric inference

KW - Single index model

UR - http://www.scopus.com/inward/record.url?scp=85012964483&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85012964483&partnerID=8YFLogxK

U2 - 10.1214/17-EJS1226

DO - 10.1214/17-EJS1226

M3 - Article

AN - SCOPUS:85012964483

VL - 11

SP - 364

EP - 384

JO - Electronic Journal of Statistics

JF - Electronic Journal of Statistics

SN - 1935-7524

IS - 1

ER -