Parameter Choice, Stability and Validity for Robust Cluster Weighted Modeling
Statistical inference based on the cluster weighted model often requires some subjective judgment from the modeler. Many features influence the final solution, such as the number of mixture components, the shape of the clusters in the explanatory variables, and the degree of heteroscedasticity of th...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-07-01
|
| Series: | Stats |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2571-905X/4/3/36 |
| _version_ | 1827681197682065408 |
|---|---|
| author | Andrea Cappozzo Luis Angel García Escudero Francesca Greselin Agustín Mayo-Iscar |
| author_facet | Andrea Cappozzo Luis Angel García Escudero Francesca Greselin Agustín Mayo-Iscar |
| author_sort | Andrea Cappozzo |
| collection | DOAJ |
| description | Statistical inference based on the cluster weighted model often requires some subjective judgment from the modeler. Many features influence the final solution, such as the number of mixture components, the shape of the clusters in the explanatory variables, and the degree of heteroscedasticity of the errors around the regression lines. Moreover, to deal with outliers and contamination that may appear in the data, hyper-parameter values ensuring robust estimation are also needed. In principle, this freedom gives rise to a variety of “legitimate” solutions, each derived by a specific set of choices and their implications in modeling. Here we introduce a method for identifying a “set of good models” to cluster a dataset, considering the whole panorama of choices. In this way, we enable the practitioner, or the scientist who needs to cluster the data, to make an educated choice. They will be able to identify the most appropriate solutions for the purposes of their own analysis, in light of their stability and validity. |
| first_indexed | 2024-03-10T07:12:47Z |
| format | Article |
| id | doaj.art-b18c1b15246f4faab532afcbf8628943 |
| institution | Directory Open Access Journal |
| issn | 2571-905X |
| language | English |
| last_indexed | 2024-03-10T07:12:47Z |
| publishDate | 2021-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Stats |
| spelling | doaj.art-b18c1b15246f4faab532afcbf86289432023-11-22T15:18:19ZengMDPI AGStats2571-905X2021-07-014360261510.3390/stats4030036Parameter Choice, Stability and Validity for Robust Cluster Weighted ModelingAndrea Cappozzo0Luis Angel García Escudero1Francesca Greselin2Agustín Mayo-Iscar3MOX-Department of Mathematics, Politecnico di Milano, 20133 Milan, ItalyDepartamento de Estadística e Investigación Operativa, Facultad de Ciencias, Universidad de Valladolid, 47002 Villadolid, SpainDepartment of Statistics and Quantitative Methods, University of Milano-Bicocca, 20126 Milan, ItalyDepartamento de Estadística e Investigación Operativa, Facultad de Ciencias, Universidad de Valladolid, 47002 Villadolid, SpainStatistical inference based on the cluster weighted model often requires some subjective judgment from the modeler. Many features influence the final solution, such as the number of mixture components, the shape of the clusters in the explanatory variables, and the degree of heteroscedasticity of the errors around the regression lines. Moreover, to deal with outliers and contamination that may appear in the data, hyper-parameter values ensuring robust estimation are also needed. In principle, this freedom gives rise to a variety of “legitimate” solutions, each derived by a specific set of choices and their implications in modeling. Here we introduce a method for identifying a “set of good models” to cluster a dataset, considering the whole panorama of choices. In this way, we enable the practitioner, or the scientist who needs to cluster the data, to make an educated choice. They will be able to identify the most appropriate solutions for the purposes of their own analysis, in light of their stability and validity.https://www.mdpi.com/2571-905X/4/3/36cluster-weighted modelingoutlierstrimmed BICeigenvalue constraintmonitoringconstrained estimation |
| spellingShingle | Andrea Cappozzo Luis Angel García Escudero Francesca Greselin Agustín Mayo-Iscar Parameter Choice, Stability and Validity for Robust Cluster Weighted Modeling Stats cluster-weighted modeling outliers trimmed BIC eigenvalue constraint monitoring constrained estimation |
| title | Parameter Choice, Stability and Validity for Robust Cluster Weighted Modeling |
| title_full | Parameter Choice, Stability and Validity for Robust Cluster Weighted Modeling |
| title_fullStr | Parameter Choice, Stability and Validity for Robust Cluster Weighted Modeling |
| title_full_unstemmed | Parameter Choice, Stability and Validity for Robust Cluster Weighted Modeling |
| title_short | Parameter Choice, Stability and Validity for Robust Cluster Weighted Modeling |
| title_sort | parameter choice stability and validity for robust cluster weighted modeling |
| topic | cluster-weighted modeling outliers trimmed BIC eigenvalue constraint monitoring constrained estimation |
| url | https://www.mdpi.com/2571-905X/4/3/36 |
| work_keys_str_mv | AT andreacappozzo parameterchoicestabilityandvalidityforrobustclusterweightedmodeling AT luisangelgarciaescudero parameterchoicestabilityandvalidityforrobustclusterweightedmodeling AT francescagreselin parameterchoicestabilityandvalidityforrobustclusterweightedmodeling AT agustinmayoiscar parameterchoicestabilityandvalidityforrobustclusterweightedmodeling |