Nonparametric generalized least squares in applied regression analysis

Michael O'Hara, Christopher Parmeter

Research output: Contribution to journalArticle

Abstract

This paper compares a nonparametric generalized least squares (NPGLS) estimator to parametric feasible GLS (FGLS) and variants of heteroscedasticity robust standard error estimators (HRSE) in an applied setting. NPGLS consistently estimates the unknown scedastic function and produces more efficient parameter estimates than HRSE. We apply these various approaches for handling heteroscedasticity to data on professor rankings obtained from RateMyProfessors.com. We find that the statistical significance of key variables differs across seven versions of HRSE, leading to different conclusions, and a standard parametric approach to FGLS suffers from misspecification. NPGLS combines the virtues of both of these parametric approaches.

Original languageEnglish (US)
Pages (from-to)456-474
Number of pages19
JournalPacific Economic Review
Volume18
Issue number4
DOIs
StatePublished - Jan 1 2013

Fingerprint

Generalized least squares
Heteroscedasticity
Regression analysis
Estimator
Standard error
Statistical significance
Ranking
Misspecification
Least squares estimator

ASJC Scopus subject areas

  • Economics and Econometrics

Cite this

Nonparametric generalized least squares in applied regression analysis. / O'Hara, Michael; Parmeter, Christopher.

In: Pacific Economic Review, Vol. 18, No. 4, 01.01.2013, p. 456-474.

Research output: Contribution to journalArticle

@article{07cf31fab99349678228785c8903c394,
title = "Nonparametric generalized least squares in applied regression analysis",
abstract = "This paper compares a nonparametric generalized least squares (NPGLS) estimator to parametric feasible GLS (FGLS) and variants of heteroscedasticity robust standard error estimators (HRSE) in an applied setting. NPGLS consistently estimates the unknown scedastic function and produces more efficient parameter estimates than HRSE. We apply these various approaches for handling heteroscedasticity to data on professor rankings obtained from RateMyProfessors.com. We find that the statistical significance of key variables differs across seven versions of HRSE, leading to different conclusions, and a standard parametric approach to FGLS suffers from misspecification. NPGLS combines the virtues of both of these parametric approaches.",
author = "Michael O'Hara and Christopher Parmeter",
year = "2013",
month = "1",
day = "1",
doi = "10.1111/1468-0106.12038",
language = "English (US)",
volume = "18",
pages = "456--474",
journal = "Pacific Economic Review",
issn = "1361-374X",
publisher = "Wiley-Blackwell",
number = "4",

}

TY - JOUR

T1 - Nonparametric generalized least squares in applied regression analysis

AU - O'Hara, Michael

AU - Parmeter, Christopher

PY - 2013/1/1

Y1 - 2013/1/1

N2 - This paper compares a nonparametric generalized least squares (NPGLS) estimator to parametric feasible GLS (FGLS) and variants of heteroscedasticity robust standard error estimators (HRSE) in an applied setting. NPGLS consistently estimates the unknown scedastic function and produces more efficient parameter estimates than HRSE. We apply these various approaches for handling heteroscedasticity to data on professor rankings obtained from RateMyProfessors.com. We find that the statistical significance of key variables differs across seven versions of HRSE, leading to different conclusions, and a standard parametric approach to FGLS suffers from misspecification. NPGLS combines the virtues of both of these parametric approaches.

AB - This paper compares a nonparametric generalized least squares (NPGLS) estimator to parametric feasible GLS (FGLS) and variants of heteroscedasticity robust standard error estimators (HRSE) in an applied setting. NPGLS consistently estimates the unknown scedastic function and produces more efficient parameter estimates than HRSE. We apply these various approaches for handling heteroscedasticity to data on professor rankings obtained from RateMyProfessors.com. We find that the statistical significance of key variables differs across seven versions of HRSE, leading to different conclusions, and a standard parametric approach to FGLS suffers from misspecification. NPGLS combines the virtues of both of these parametric approaches.

UR - http://www.scopus.com/inward/record.url?scp=84885695564&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84885695564&partnerID=8YFLogxK

U2 - 10.1111/1468-0106.12038

DO - 10.1111/1468-0106.12038

M3 - Article

VL - 18

SP - 456

EP - 474

JO - Pacific Economic Review

JF - Pacific Economic Review

SN - 1361-374X

IS - 4

ER -