Computationally simple estimation and improved efficiency for special cases of double truncation

Matthew D. Austin, David K. Simon, Rebecca Betensky

Research output: Contribution to journalArticle

Abstract

Doubly truncated survival data arise when event times are observed only if they occur within subject specific intervals of times. Existing iterative estimation procedures for doubly truncated data are computationally intensive (Turnbull 38:290-295, 1976; Efron and Petrosian 94:824-825, 1999; Shen 62:835-853, 2010a). These procedures assume that the event time is independent of the truncation times, in the sample space that conforms to their requisite ordering. This type of independence is referred to as quasi-independence. In this paper we identify and consider two special cases of quasi-independence: complete quasi-independence and complete truncation dependence. For the case of complete quasi-independence, we derive the nonparametric maximum likelihood estimator in closed-form. For the case of complete truncation dependence, we derive a closed-form nonparametric estimator that requires some external information, and a semi-parametric maximum likelihood estimator that achieves improved efficiency relative to the standard nonparametric maximum likelihood estimator, in the absence of external information. We demonstrate the consistency and potentially improved efficiency of the estimators in simulation studies, and illustrate their use in application to studies of AIDS incubation and Parkinson's disease age of onset.

Original languageEnglish (US)
Pages (from-to)335-354
Number of pages20
JournalLifetime Data Analysis
Volume20
Issue number3
DOIs
StatePublished - Jan 1 2014

Fingerprint

Quasi-independence
Truncation
Maximum likelihood
Nonparametric Maximum Likelihood Estimator
Truncated Data
Closed-form
Parkinson's Disease
Sample space
Relative Efficiency
Survival Data
Nonparametric Estimator
Age of Onset
Maximum Likelihood Estimator
Parkinson Disease
Acquired Immunodeficiency Syndrome
Simulation Study
Estimator
Interval
Demonstrate

Keywords

  • Piecewise exponential
  • Quasi-independence
  • Semi-parametric
  • Survival analysis

ASJC Scopus subject areas

  • Applied Mathematics
  • Medicine(all)

Cite this

Computationally simple estimation and improved efficiency for special cases of double truncation. / Austin, Matthew D.; Simon, David K.; Betensky, Rebecca.

In: Lifetime Data Analysis, Vol. 20, No. 3, 01.01.2014, p. 335-354.

Research output: Contribution to journalArticle

Austin, Matthew D. ; Simon, David K. ; Betensky, Rebecca. / Computationally simple estimation and improved efficiency for special cases of double truncation. In: Lifetime Data Analysis. 2014 ; Vol. 20, No. 3. pp. 335-354.
@article{6adc717f318444ec9e8aec5330ce955f,
title = "Computationally simple estimation and improved efficiency for special cases of double truncation",
abstract = "Doubly truncated survival data arise when event times are observed only if they occur within subject specific intervals of times. Existing iterative estimation procedures for doubly truncated data are computationally intensive (Turnbull 38:290-295, 1976; Efron and Petrosian 94:824-825, 1999; Shen 62:835-853, 2010a). These procedures assume that the event time is independent of the truncation times, in the sample space that conforms to their requisite ordering. This type of independence is referred to as quasi-independence. In this paper we identify and consider two special cases of quasi-independence: complete quasi-independence and complete truncation dependence. For the case of complete quasi-independence, we derive the nonparametric maximum likelihood estimator in closed-form. For the case of complete truncation dependence, we derive a closed-form nonparametric estimator that requires some external information, and a semi-parametric maximum likelihood estimator that achieves improved efficiency relative to the standard nonparametric maximum likelihood estimator, in the absence of external information. We demonstrate the consistency and potentially improved efficiency of the estimators in simulation studies, and illustrate their use in application to studies of AIDS incubation and Parkinson's disease age of onset.",
keywords = "Piecewise exponential, Quasi-independence, Semi-parametric, Survival analysis",
author = "Austin, {Matthew D.} and Simon, {David K.} and Rebecca Betensky",
year = "2014",
month = "1",
day = "1",
doi = "10.1007/s10985-013-9287-z",
language = "English (US)",
volume = "20",
pages = "335--354",
journal = "Lifetime Data Analysis",
issn = "1380-7870",
publisher = "Springer Netherlands",
number = "3",

}

TY - JOUR

T1 - Computationally simple estimation and improved efficiency for special cases of double truncation

AU - Austin, Matthew D.

AU - Simon, David K.

AU - Betensky, Rebecca

PY - 2014/1/1

Y1 - 2014/1/1

N2 - Doubly truncated survival data arise when event times are observed only if they occur within subject specific intervals of times. Existing iterative estimation procedures for doubly truncated data are computationally intensive (Turnbull 38:290-295, 1976; Efron and Petrosian 94:824-825, 1999; Shen 62:835-853, 2010a). These procedures assume that the event time is independent of the truncation times, in the sample space that conforms to their requisite ordering. This type of independence is referred to as quasi-independence. In this paper we identify and consider two special cases of quasi-independence: complete quasi-independence and complete truncation dependence. For the case of complete quasi-independence, we derive the nonparametric maximum likelihood estimator in closed-form. For the case of complete truncation dependence, we derive a closed-form nonparametric estimator that requires some external information, and a semi-parametric maximum likelihood estimator that achieves improved efficiency relative to the standard nonparametric maximum likelihood estimator, in the absence of external information. We demonstrate the consistency and potentially improved efficiency of the estimators in simulation studies, and illustrate their use in application to studies of AIDS incubation and Parkinson's disease age of onset.

AB - Doubly truncated survival data arise when event times are observed only if they occur within subject specific intervals of times. Existing iterative estimation procedures for doubly truncated data are computationally intensive (Turnbull 38:290-295, 1976; Efron and Petrosian 94:824-825, 1999; Shen 62:835-853, 2010a). These procedures assume that the event time is independent of the truncation times, in the sample space that conforms to their requisite ordering. This type of independence is referred to as quasi-independence. In this paper we identify and consider two special cases of quasi-independence: complete quasi-independence and complete truncation dependence. For the case of complete quasi-independence, we derive the nonparametric maximum likelihood estimator in closed-form. For the case of complete truncation dependence, we derive a closed-form nonparametric estimator that requires some external information, and a semi-parametric maximum likelihood estimator that achieves improved efficiency relative to the standard nonparametric maximum likelihood estimator, in the absence of external information. We demonstrate the consistency and potentially improved efficiency of the estimators in simulation studies, and illustrate their use in application to studies of AIDS incubation and Parkinson's disease age of onset.

KW - Piecewise exponential

KW - Quasi-independence

KW - Semi-parametric

KW - Survival analysis

UR - http://www.scopus.com/inward/record.url?scp=84902381253&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84902381253&partnerID=8YFLogxK

U2 - 10.1007/s10985-013-9287-z

DO - 10.1007/s10985-013-9287-z

M3 - Article

VL - 20

SP - 335

EP - 354

JO - Lifetime Data Analysis

JF - Lifetime Data Analysis

SN - 1380-7870

IS - 3

ER -