Number of the records: 1
A Robustified Metalearning Procedure for Regression Estimators
- 1.
SYSNO ASEP 0510554 Document Type C - Proceedings Paper (int. conf.) R&D Document Type Conference Paper Title A Robustified Metalearning Procedure for Regression Estimators Author(s) Kalina, Jan (UIVT-O) RID, SAI, ORCID
Neoral, A. (CZ)Source Title The 13th International Days of Statistics and Economics Conference Proceedings. - Slaný : Melandrium, 2019 / Löster T. ; Pavelka T. - ISBN 978-80-87990-18-6 Pages s. 617-626 Number of pages 10 s. Publication form Online - E Action International Days of Statistics and Economics /13./ Event date 05.09.2019 - 07.09.2019 VEvent location Prague Country CZ - Czech Republic Event type WRD Language eng - English Country CZ - Czech Republic Keywords model choice ; computational statistics ; robustness ; variable selection Subject RIV IN - Informatics, Computer Science OECD category Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8) Institutional support UIVT-O - RVO:67985807 UT WOS 000589182000062 DOI 10.18267/pr.2019.los.186.61 Annotation 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. Workplace Institute of Computer Science Contact Tereza Šírová, sirova@cs.cas.cz, Tel.: 266 053 800 Year of Publishing 2020 Electronic address https://msed.vse.cz/msed_2019/sbornik/toc.html
Number of the records: 1