Trimmed and winsorized semiparametric estimator for left-truncated and right-censored regression models

Myoung-jae Lee, Maria Karlsson

Research output: Contribution to journalArticle

Abstract

For a linear regression model subject to left-truncation and right-censoring where the truncation and censoring points are known constants (or always observed if random), Karlsson and Laitila (Stat Probab Lett 78:2567–2571, 2008) proposed a semiparametric estimator which deals with left-truncation by trimming and right-censoring by ‘winsorizing’. The estimator was motivated by a zero moment condition where a transformed error term appears with trimmed and winsorized tails. This paper takes the semiparametric estimator further by deriving the asymptotic distribution that was not shown in Karlsson and Laitila (Stat Probab Lett 78:2567–2571, 2008) and discusses its implementation aspects in practice, albeit brief.

Original languageEnglish
Pages (from-to)485-495
Number of pages11
JournalMetrika
Volume78
Issue number4
DOIs
Publication statusPublished - 2015 May 1

Fingerprint

Censored Regression
Left Truncation
Right Censoring
Regression Model
Estimator
Trimming
Moment Conditions
Censoring
Error term
Linear Regression Model
Truncation
Asymptotic distribution
Tail
Zero
Semiparametric estimators
Censored regression model

Keywords

  • Left truncation
  • LTRC
  • Right censoring
  • Semiparametrics
  • Trimmed mean
  • Winsorized mean

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

Trimmed and winsorized semiparametric estimator for left-truncated and right-censored regression models. / Lee, Myoung-jae; Karlsson, Maria.

In: Metrika, Vol. 78, No. 4, 01.05.2015, p. 485-495.

Research output: Contribution to journalArticle

@article{d796e683ff994f03892744780019fbfc,
title = "Trimmed and winsorized semiparametric estimator for left-truncated and right-censored regression models",
abstract = "For a linear regression model subject to left-truncation and right-censoring where the truncation and censoring points are known constants (or always observed if random), Karlsson and Laitila (Stat Probab Lett 78:2567–2571, 2008) proposed a semiparametric estimator which deals with left-truncation by trimming and right-censoring by ‘winsorizing’. The estimator was motivated by a zero moment condition where a transformed error term appears with trimmed and winsorized tails. This paper takes the semiparametric estimator further by deriving the asymptotic distribution that was not shown in Karlsson and Laitila (Stat Probab Lett 78:2567–2571, 2008) and discusses its implementation aspects in practice, albeit brief.",
keywords = "Left truncation, LTRC, Right censoring, Semiparametrics, Trimmed mean, Winsorized mean",
author = "Myoung-jae Lee and Maria Karlsson",
year = "2015",
month = "5",
day = "1",
doi = "10.1007/s00184-014-0513-9",
language = "English",
volume = "78",
pages = "485--495",
journal = "Metrika",
issn = "0026-1335",
publisher = "Springer Verlag",
number = "4",

}

TY - JOUR

T1 - Trimmed and winsorized semiparametric estimator for left-truncated and right-censored regression models

AU - Lee, Myoung-jae

AU - Karlsson, Maria

PY - 2015/5/1

Y1 - 2015/5/1

N2 - For a linear regression model subject to left-truncation and right-censoring where the truncation and censoring points are known constants (or always observed if random), Karlsson and Laitila (Stat Probab Lett 78:2567–2571, 2008) proposed a semiparametric estimator which deals with left-truncation by trimming and right-censoring by ‘winsorizing’. The estimator was motivated by a zero moment condition where a transformed error term appears with trimmed and winsorized tails. This paper takes the semiparametric estimator further by deriving the asymptotic distribution that was not shown in Karlsson and Laitila (Stat Probab Lett 78:2567–2571, 2008) and discusses its implementation aspects in practice, albeit brief.

AB - For a linear regression model subject to left-truncation and right-censoring where the truncation and censoring points are known constants (or always observed if random), Karlsson and Laitila (Stat Probab Lett 78:2567–2571, 2008) proposed a semiparametric estimator which deals with left-truncation by trimming and right-censoring by ‘winsorizing’. The estimator was motivated by a zero moment condition where a transformed error term appears with trimmed and winsorized tails. This paper takes the semiparametric estimator further by deriving the asymptotic distribution that was not shown in Karlsson and Laitila (Stat Probab Lett 78:2567–2571, 2008) and discusses its implementation aspects in practice, albeit brief.

KW - Left truncation

KW - LTRC

KW - Right censoring

KW - Semiparametrics

KW - Trimmed mean

KW - Winsorized mean

UR - http://www.scopus.com/inward/record.url?scp=84940008309&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84940008309&partnerID=8YFLogxK

U2 - 10.1007/s00184-014-0513-9

DO - 10.1007/s00184-014-0513-9

M3 - Article

AN - SCOPUS:84940008309

VL - 78

SP - 485

EP - 495

JO - Metrika

JF - Metrika

SN - 0026-1335

IS - 4

ER -