Nearly weighted risk minimal unbiased estimation

Ulrich K. Müller, Yulong Wang

Research output: Contribution to journalArticle

Abstract

Consider a small-sample parametric estimation problem, such as the estimation of the coefficient in a Gaussian AR(1). We develop a numerical algorithm that determines an estimator that is nearly (mean or median) unbiased, and among all such estimators, comes close to minimizing a weighted average risk criterion. We also apply our generic approach to the median unbiased estimation of the degree of time variation in a Gaussian local-level model, and to a quantile unbiased point forecast for a Gaussian AR(1) process.

Original languageEnglish (US)
Pages (from-to)18-34
Number of pages17
JournalJournal of Econometrics
Volume209
Issue number1
DOIs
StatePublished - Mar 1 2019

Fingerprint

Median
Estimator
Time variation
Numerical algorithms
Coefficients
Point forecasts
Quantile
Small sample

Keywords

  • Autoregression
  • Mean bias
  • Median bias
  • Quantile unbiased forecast

ASJC Scopus subject areas

  • Economics and Econometrics

Cite this

Nearly weighted risk minimal unbiased estimation. / Müller, Ulrich K.; Wang, Yulong.

In: Journal of Econometrics, Vol. 209, No. 1, 01.03.2019, p. 18-34.

Research output: Contribution to journalArticle

Müller, Ulrich K. ; Wang, Yulong. / Nearly weighted risk minimal unbiased estimation. In: Journal of Econometrics. 2019 ; Vol. 209, No. 1. pp. 18-34.
@article{0e8f9244277a442c999dd7a97f232dac,
title = "Nearly weighted risk minimal unbiased estimation",
abstract = "Consider a small-sample parametric estimation problem, such as the estimation of the coefficient in a Gaussian AR(1). We develop a numerical algorithm that determines an estimator that is nearly (mean or median) unbiased, and among all such estimators, comes close to minimizing a weighted average risk criterion. We also apply our generic approach to the median unbiased estimation of the degree of time variation in a Gaussian local-level model, and to a quantile unbiased point forecast for a Gaussian AR(1) process.",
keywords = "Autoregression, Mean bias, Median bias, Quantile unbiased forecast",
author = "M{\"u}ller, {Ulrich K.} and Yulong Wang",
year = "2019",
month = "3",
day = "1",
doi = "10.1016/j.jeconom.2018.11.016",
language = "English (US)",
volume = "209",
pages = "18--34",
journal = "Journal of Econometrics",
issn = "0304-4076",
publisher = "Elsevier",
number = "1",

}

TY - JOUR

T1 - Nearly weighted risk minimal unbiased estimation

AU - Müller, Ulrich K.

AU - Wang, Yulong

PY - 2019/3/1

Y1 - 2019/3/1

N2 - Consider a small-sample parametric estimation problem, such as the estimation of the coefficient in a Gaussian AR(1). We develop a numerical algorithm that determines an estimator that is nearly (mean or median) unbiased, and among all such estimators, comes close to minimizing a weighted average risk criterion. We also apply our generic approach to the median unbiased estimation of the degree of time variation in a Gaussian local-level model, and to a quantile unbiased point forecast for a Gaussian AR(1) process.

AB - Consider a small-sample parametric estimation problem, such as the estimation of the coefficient in a Gaussian AR(1). We develop a numerical algorithm that determines an estimator that is nearly (mean or median) unbiased, and among all such estimators, comes close to minimizing a weighted average risk criterion. We also apply our generic approach to the median unbiased estimation of the degree of time variation in a Gaussian local-level model, and to a quantile unbiased point forecast for a Gaussian AR(1) process.

KW - Autoregression

KW - Mean bias

KW - Median bias

KW - Quantile unbiased forecast

UR - http://www.scopus.com/inward/record.url?scp=85059248051&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85059248051&partnerID=8YFLogxK

U2 - 10.1016/j.jeconom.2018.11.016

DO - 10.1016/j.jeconom.2018.11.016

M3 - Article

VL - 209

SP - 18

EP - 34

JO - Journal of Econometrics

JF - Journal of Econometrics

SN - 0304-4076

IS - 1

ER -