Bias reduction by imputation for linear panel data models with nonrandom missing

Goeun Lee, Chirok Han

Research output: Contribution to journalArticle

Abstract

When no variables are observed for endogenous non-respondents of panel data, bias correction is available only for a limited class of instrumental variable estimators, which require strong conditions for consistency and often suffer from substantial efficiency loss. In this paper we examine a convenient alternative method of imputing the missing explanatory variables and then using standard bias-correction procedures for sample selection. Various bias-corrected estimators are derived and their performances are compared by Monte Carlo experiments. Results verify efficiency loss by the instrumental variable estimators and suggest that the imputation method is practically useful if it is applied to first-difference regression.

Original languageEnglish
Pages (from-to)1-25
Number of pages25
JournalJournal of Economic Theory and Econometrics
Volume29
Issue number1
Publication statusPublished - 2018 Mar 1

Fingerprint

Instrumental variables estimator
Bias correction
Bias reduction
Imputation
Estimator
Panel data
Monte Carlo experiment
Sample selection

Keywords

  • Attrition
  • Bias-correction
  • Imputation
  • Missing
  • Nonresponse
  • Panel data
  • Selection

ASJC Scopus subject areas

  • Economics and Econometrics

Cite this

Bias reduction by imputation for linear panel data models with nonrandom missing. / Lee, Goeun; Han, Chirok.

In: Journal of Economic Theory and Econometrics, Vol. 29, No. 1, 01.03.2018, p. 1-25.

Research output: Contribution to journalArticle

@article{aae015895d9c46869fc07438ae74fbcb,
title = "Bias reduction by imputation for linear panel data models with nonrandom missing",
abstract = "When no variables are observed for endogenous non-respondents of panel data, bias correction is available only for a limited class of instrumental variable estimators, which require strong conditions for consistency and often suffer from substantial efficiency loss. In this paper we examine a convenient alternative method of imputing the missing explanatory variables and then using standard bias-correction procedures for sample selection. Various bias-corrected estimators are derived and their performances are compared by Monte Carlo experiments. Results verify efficiency loss by the instrumental variable estimators and suggest that the imputation method is practically useful if it is applied to first-difference regression.",
keywords = "Attrition, Bias-correction, Imputation, Missing, Nonresponse, Panel data, Selection",
author = "Goeun Lee and Chirok Han",
year = "2018",
month = "3",
day = "1",
language = "English",
volume = "29",
pages = "1--25",
journal = "Journal of Economic Theory and Econometrics",
issn = "1229-2893",
publisher = "Korean Econometric Society",
number = "1",

}

TY - JOUR

T1 - Bias reduction by imputation for linear panel data models with nonrandom missing

AU - Lee, Goeun

AU - Han, Chirok

PY - 2018/3/1

Y1 - 2018/3/1

N2 - When no variables are observed for endogenous non-respondents of panel data, bias correction is available only for a limited class of instrumental variable estimators, which require strong conditions for consistency and often suffer from substantial efficiency loss. In this paper we examine a convenient alternative method of imputing the missing explanatory variables and then using standard bias-correction procedures for sample selection. Various bias-corrected estimators are derived and their performances are compared by Monte Carlo experiments. Results verify efficiency loss by the instrumental variable estimators and suggest that the imputation method is practically useful if it is applied to first-difference regression.

AB - When no variables are observed for endogenous non-respondents of panel data, bias correction is available only for a limited class of instrumental variable estimators, which require strong conditions for consistency and often suffer from substantial efficiency loss. In this paper we examine a convenient alternative method of imputing the missing explanatory variables and then using standard bias-correction procedures for sample selection. Various bias-corrected estimators are derived and their performances are compared by Monte Carlo experiments. Results verify efficiency loss by the instrumental variable estimators and suggest that the imputation method is practically useful if it is applied to first-difference regression.

KW - Attrition

KW - Bias-correction

KW - Imputation

KW - Missing

KW - Nonresponse

KW - Panel data

KW - Selection

UR - http://www.scopus.com/inward/record.url?scp=85046549531&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85046549531&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:85046549531

VL - 29

SP - 1

EP - 25

JO - Journal of Economic Theory and Econometrics

JF - Journal of Economic Theory and Econometrics

SN - 1229-2893

IS - 1

ER -