A Robustified Metalearning Procedure for Regression Estimators

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F67985556%3A_____%2F19%3A00510771" target="_blank" >RIV/67985556:_____/19:00510771 - isvavai.cz</a>

Nalezeny alternativní kódy

RIV/67985807:_____/19:00510554

Výsledek na webu

<a href="http://dx.doi.org/10.18267/pr.2019.los.186.61" target="_blank" >http://dx.doi.org/10.18267/pr.2019.los.186.61</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.18267/pr.2019.los.186.61" target="_blank" >10.18267/pr.2019.los.186.61</a>

Jazyk výsledku

angličtina

Název v původním jazyce

A Robustified Metalearning Procedure for Regression Estimators

Popis výsledku v původním jazyce

Metalearning represents a useful methodology for selecting and recommending a suitable algorithm or method for a new dataset exploiting a database of training datasets. While metalearning is potentially beneficial for the analysis of economic data, we must be aware of its instability and sensitivity to outlying measurements (outliers) as well as measurement errors. The aim of this paper is to robustify the metalearning process. First, we prepare some useful theoretical tools exploiting the idea of implicit weighting, inspired by the least weighted squares estimator. These include a robust coefficient of determination, a robust version of mean square error, and a simple rule for outlier detection in linear regression. We perform a metalearning study for recommending the best linear regression estimator for a new dataset (not included in the training database). The prediction of the optimal estimator is learned over a set of 20 real datasets with economic motivation, while the least squares are compared with several (highly) robust estimators. We investigate the effect of variable selection on the metalearning results. If the training as well as validation data are considered after a proper robust variable selection, the metalearning performance is improved remarkably, especially if a robust prediction error is used.

Název v anglickém jazyce

A Robustified Metalearning Procedure for Regression Estimators

Popis výsledku anglicky

Metalearning represents a useful methodology for selecting and recommending a suitable algorithm or method for a new dataset exploiting a database of training datasets. While metalearning is potentially beneficial for the analysis of economic data, we must be aware of its instability and sensitivity to outlying measurements (outliers) as well as measurement errors. The aim of this paper is to robustify the metalearning process. First, we prepare some useful theoretical tools exploiting the idea of implicit weighting, inspired by the least weighted squares estimator. These include a robust coefficient of determination, a robust version of mean square error, and a simple rule for outlier detection in linear regression. We perform a metalearning study for recommending the best linear regression estimator for a new dataset (not included in the training database). The prediction of the optimal estimator is learned over a set of 20 real datasets with economic motivation, while the least squares are compared with several (highly) robust estimators. We investigate the effect of variable selection on the metalearning results. If the training as well as validation data are considered after a proper robust variable selection, the metalearning performance is improved remarkably, especially if a robust prediction error is used.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10101 - Pure mathematics

Projekt

<a href="/cs/project/GA17-07384S" target="_blank" >GA17-07384S: Neparametrické (statistické) metody v moderní ekonometrii</a><br>

Návaznosti

I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Rok uplatnění

2019

Kód důvěrnosti údajů

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

Název statě ve sborníku

The 13th International Days of Statistics and Economics Conference Proceedings

ISBN

978-80-87990-18-6

ISSN

—

e-ISSN

—

Počet stran výsledku

10

Strana od-do

617-626

Název nakladatele

Melandrium

Místo vydání

Slaný

Místo konání akce

Prague

Datum konání akce

5. 9. 2019

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—