Robust Model Selection Criteria Based on Pseudodistances
In this paper, we introduce a new class of robust model selection criteria. These criteria are defined by estimators of the expected overall discrepancy using pseudodistances and the minimum pseudodistance principle. Theoretical properties of these criteria are proved, namely asymptotic unbiasedness...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2020-03-01
|
Series: | Entropy |
Subjects: | |
Online Access: | https://www.mdpi.com/1099-4300/22/3/304 |
_version_ | 1811304885911552000 |
---|---|
author | Aida Toma Alex Karagrigoriou Paschalini Trentou |
author_facet | Aida Toma Alex Karagrigoriou Paschalini Trentou |
author_sort | Aida Toma |
collection | DOAJ |
description | In this paper, we introduce a new class of robust model selection criteria. These criteria are defined by estimators of the expected overall discrepancy using pseudodistances and the minimum pseudodistance principle. Theoretical properties of these criteria are proved, namely asymptotic unbiasedness, robustness, consistency, as well as the limit laws. The case of the linear regression models is studied and a specific pseudodistance based criterion is proposed. Monte Carlo simulations and applications for real data are presented in order to exemplify the performance of the new methodology. These examples show that the new selection criterion for regression models is a good competitor of some well known criteria and may have superior performance, especially in the case of small and contaminated samples. |
first_indexed | 2024-04-13T08:15:25Z |
format | Article |
id | doaj.art-15b3560deb2945bd944aac98e39f088f |
institution | Directory Open Access Journal |
issn | 1099-4300 |
language | English |
last_indexed | 2024-04-13T08:15:25Z |
publishDate | 2020-03-01 |
publisher | MDPI AG |
record_format | Article |
series | Entropy |
spelling | doaj.art-15b3560deb2945bd944aac98e39f088f2022-12-22T02:54:48ZengMDPI AGEntropy1099-43002020-03-0122330410.3390/e22030304e22030304Robust Model Selection Criteria Based on PseudodistancesAida Toma0Alex Karagrigoriou1Paschalini Trentou2Department of Applied Mathematics, Bucharest University of Economic Studies, 010164 Bucharest, RomaniaDepartment of Statistics and Actuarial-Financial Mathematics, Lab of Statistics and Data Analysis, University of the Aegean, 83200 Karlovasi, GreeceDepartment of Statistics and Actuarial-Financial Mathematics, Lab of Statistics and Data Analysis, University of the Aegean, 83200 Karlovasi, GreeceIn this paper, we introduce a new class of robust model selection criteria. These criteria are defined by estimators of the expected overall discrepancy using pseudodistances and the minimum pseudodistance principle. Theoretical properties of these criteria are proved, namely asymptotic unbiasedness, robustness, consistency, as well as the limit laws. The case of the linear regression models is studied and a specific pseudodistance based criterion is proposed. Monte Carlo simulations and applications for real data are presented in order to exemplify the performance of the new methodology. These examples show that the new selection criterion for regression models is a good competitor of some well known criteria and may have superior performance, especially in the case of small and contaminated samples.https://www.mdpi.com/1099-4300/22/3/304model selectionminimum pseudodistance estimationrobustness |
spellingShingle | Aida Toma Alex Karagrigoriou Paschalini Trentou Robust Model Selection Criteria Based on Pseudodistances Entropy model selection minimum pseudodistance estimation robustness |
title | Robust Model Selection Criteria Based on Pseudodistances |
title_full | Robust Model Selection Criteria Based on Pseudodistances |
title_fullStr | Robust Model Selection Criteria Based on Pseudodistances |
title_full_unstemmed | Robust Model Selection Criteria Based on Pseudodistances |
title_short | Robust Model Selection Criteria Based on Pseudodistances |
title_sort | robust model selection criteria based on pseudodistances |
topic | model selection minimum pseudodistance estimation robustness |
url | https://www.mdpi.com/1099-4300/22/3/304 |
work_keys_str_mv | AT aidatoma robustmodelselectioncriteriabasedonpseudodistances AT alexkaragrigoriou robustmodelselectioncriteriabasedonpseudodistances AT paschalinitrentou robustmodelselectioncriteriabasedonpseudodistances |