Interdependency Pattern Recognition in Econometrics: A Penalized Regularization Antidote
When it comes to variable interpretation, multicollinearity is among the biggest issues that must be surmounted, especially in this new era of Big Data Analytics. Since even moderate size multicollinearity can prevent proper interpretation, special diagnostics must be recommended and implemented for...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-12-01
|
| Series: | Econometrics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2225-1146/9/4/44 |
| _version_ | 1797505313516027904 |
|---|---|
| author | Kimon Ntotsis Alex Karagrigoriou Andreas Artemiou |
| author_facet | Kimon Ntotsis Alex Karagrigoriou Andreas Artemiou |
| author_sort | Kimon Ntotsis |
| collection | DOAJ |
| description | When it comes to variable interpretation, multicollinearity is among the biggest issues that must be surmounted, especially in this new era of Big Data Analytics. Since even moderate size multicollinearity can prevent proper interpretation, special diagnostics must be recommended and implemented for identification purposes. Nonetheless, in the areas of econometrics and statistics, among other fields, these diagnostics are controversial concerning their “successfulness”. It has been remarked that they frequently fail to do proper model assessment due to information complexity, resulting in model misspecification. This work proposes and investigates a robust and easily interpretable methodology, termed Elastic Information Criterion, capable of capturing multicollinearity rather accurately and effectively and thus providing a proper model assessment. The performance is investigated via simulated and real data. |
| first_indexed | 2024-03-10T04:16:50Z |
| format | Article |
| id | doaj.art-44c92730dcdb44cca2e9525a506d5e2d |
| institution | Directory Open Access Journal |
| issn | 2225-1146 |
| language | English |
| last_indexed | 2024-03-10T04:16:50Z |
| publishDate | 2021-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Econometrics |
| spelling | doaj.art-44c92730dcdb44cca2e9525a506d5e2d2023-11-23T07:58:21ZengMDPI AGEconometrics2225-11462021-12-01944410.3390/econometrics9040044Interdependency Pattern Recognition in Econometrics: A Penalized Regularization AntidoteKimon Ntotsis0Alex Karagrigoriou1Andreas Artemiou2Department of Statistics and Actuarial-Financial Mathematics, University of the Aegean, 832 00 Karlovasi, GreeceDepartment of Statistics and Actuarial-Financial Mathematics, University of the Aegean, 832 00 Karlovasi, GreeceSchool of Mathematics, Cardiff University, Cardiff CF10 3AT, UKWhen it comes to variable interpretation, multicollinearity is among the biggest issues that must be surmounted, especially in this new era of Big Data Analytics. Since even moderate size multicollinearity can prevent proper interpretation, special diagnostics must be recommended and implemented for identification purposes. Nonetheless, in the areas of econometrics and statistics, among other fields, these diagnostics are controversial concerning their “successfulness”. It has been remarked that they frequently fail to do proper model assessment due to information complexity, resulting in model misspecification. This work proposes and investigates a robust and easily interpretable methodology, termed Elastic Information Criterion, capable of capturing multicollinearity rather accurately and effectively and thus providing a proper model assessment. The performance is investigated via simulated and real data.https://www.mdpi.com/2225-1146/9/4/44multicollinearity detectionElastic Net Regularizationfeature selectiondimensionality reductioncoefficient penalization |
| spellingShingle | Kimon Ntotsis Alex Karagrigoriou Andreas Artemiou Interdependency Pattern Recognition in Econometrics: A Penalized Regularization Antidote Econometrics multicollinearity detection Elastic Net Regularization feature selection dimensionality reduction coefficient penalization |
| title | Interdependency Pattern Recognition in Econometrics: A Penalized Regularization Antidote |
| title_full | Interdependency Pattern Recognition in Econometrics: A Penalized Regularization Antidote |
| title_fullStr | Interdependency Pattern Recognition in Econometrics: A Penalized Regularization Antidote |
| title_full_unstemmed | Interdependency Pattern Recognition in Econometrics: A Penalized Regularization Antidote |
| title_short | Interdependency Pattern Recognition in Econometrics: A Penalized Regularization Antidote |
| title_sort | interdependency pattern recognition in econometrics a penalized regularization antidote |
| topic | multicollinearity detection Elastic Net Regularization feature selection dimensionality reduction coefficient penalization |
| url | https://www.mdpi.com/2225-1146/9/4/44 |
| work_keys_str_mv | AT kimonntotsis interdependencypatternrecognitionineconometricsapenalizedregularizationantidote AT alexkaragrigoriou interdependencypatternrecognitionineconometricsapenalizedregularizationantidote AT andreasartemiou interdependencypatternrecognitionineconometricsapenalizedregularizationantidote |