On Robust Testing for Normality of Error Terms in Regression Models

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F62156489%3A43110%2F15%3A43906564" target="_blank" >RIV/62156489:43110/15:43906564 - isvavai.cz</a>

Výsledek na webu

<a href="http://mme2015.zcu.cz/downloads/MME_2015_proceedings.pdf" target="_blank" >http://mme2015.zcu.cz/downloads/MME_2015_proceedings.pdf</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

On Robust Testing for Normality of Error Terms in Regression Models

Popis výsledku v původním jazyce

Testing for normality of error terms constitutes one of the most important steps of regression model verification and validation, because failure to assess non-normality of the regression residuals may lead to incorrect results since significant deviations from normality can substantially affect the performance of usual statistical inference techniques. Thus, in the majority of cases of relevant regression analysis normality of error terms is expected. However, this is not true in many practical situations. While OLS estimator is known to be very sensitive to outliers, the robust regression estimator (e.g. Least Trimmed Squares, LTS) is known not to be unduly affected by the presence of outliers. The aim of this paper is to present and discuss the trade-off between power and robustness of selected classical and robust normality tests of error terms in regression models. For this purpose we use OLS and LTS residuals from linear regression models with various distributed dependent variab

Název v anglickém jazyce

On Robust Testing for Normality of Error Terms in Regression Models

Popis výsledku anglicky

Testing for normality of error terms constitutes one of the most important steps of regression model verification and validation, because failure to assess non-normality of the regression residuals may lead to incorrect results since significant deviations from normality can substantially affect the performance of usual statistical inference techniques. Thus, in the majority of cases of relevant regression analysis normality of error terms is expected. However, this is not true in many practical situations. While OLS estimator is known to be very sensitive to outliers, the robust regression estimator (e.g. Least Trimmed Squares, LTS) is known not to be unduly affected by the presence of outliers. The aim of this paper is to present and discuss the trade-off between power and robustness of selected classical and robust normality tests of error terms in regression models. For this purpose we use OLS and LTS residuals from linear regression models with various distributed dependent variab

Druh

D - Stať ve sborníku

CEP obor

BB - Aplikovaná statistika, operační výzkum

OECD FORD obor

—

Projekt

—

Návaznosti

I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Rok uplatnění

2015

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Mathematical Methods in Economics 2015: Conference Proceedings

ISBN

978-80-261-0539-8

ISSN

—

e-ISSN

—

Počet stran výsledku

6

Strana od-do

749-754

Název nakladatele

Západočeská univerzita

Místo vydání

Plzeň

Místo konání akce

Cheb

Datum konání akce

9. 9. 2015

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—