Robust Metalearning: Comparing Robust Regression Using A Robust Prediction Error

0497292 - ÚI 2019 RIV CZ eng C - Conference Paper (international conference)
Peštová, Barbora - Kalina, Jan
Robust Metalearning: Comparing Robust Regression Using A Robust Prediction Error.
The 12th International Days of Statistics and Economics Conference Proceedings. Slaný: Melandrium, 2018 - (Löster, T.; Pavelka, T.), s. 1367-1376. ISBN 978-80-87990-14-8.
[International Days of Statistics and Economics /12./. Prague (CZ), 06.09.2018-08.09.2018]
R&D Projects: GA ČR GA17-01251S
Institutional support: RVO:67985807
Keywords : metalearning * robust regression * outliers * robust prediction error
OECD category: Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)
https://msed.vse.cz/msed_2018/article/13-Pestova-Barbora-paper.pdf

The aim of this paper is to construct a classification rule for predicting the best regression estimator for a new data set based on a database of 20 training data sets. Various estimators considered here include some popular methods of robust statistics. The methodology used for constructing the classification rule can be described as metalearning. Nevertheless, standard approaches of metalearning should be robustified if working with data sets contaminated by outlying measurements (outliers). Therefore, our contribution can be also described as robustification of the metalearning process by using a robust prediction error. In addition to performing the metalearning study by means of both standard and robust approaches, we search for a detailed interpretation in two particular situations. The results of detailed investigation show that the knowledge obtained by a metalearning approach standing on standard principles is prone to great variability and instability, which makes it hard to believe that the results are not just a consequence of a mere chance. Such aspect of metalearning seems not to have been previously analyzed in literature.
Permanent Link: http://hdl.handle.net/11104/0289877