Robust Clustering of Banks in Argentina
The purpose of this paper is to classify and characterize 64 banks, active as of 2010 in Argentina, by means of robust techniques used on information gathered during the period 2001-2010. Based on the strategy criteria established in (Wang 2007) and (Werbin 2010), seven variables were selected. In a...
| Main Authors: | , , |
|---|---|
| Format: | Online |
| Language: | eng |
| Published: |
Instituto de Economía y Finanzas. Facultad de Ciencias Económicas. Universidada Nacional de Córdoba.
2018
|
| Subjects: | |
| Online Access: | https://revistas.unc.edu.ar/index.php/REyE/article/view/29385 |
| _version_ | 1839026225264197632 |
|---|---|
| author | Vargas, José M. Díaz, Margarita García, Fernando Vargas, José M. Díaz, Margarita García, Fernando |
| author_facet | Vargas, José M. Díaz, Margarita García, Fernando Vargas, José M. Díaz, Margarita García, Fernando |
| author_sort | Vargas, José M. |
| collection | Portal de Revistas |
| description | The purpose of this paper is to classify and characterize 64 banks, active as of 2010 in Argentina, by means of robust techniques used on information gathered during the period 2001-2010. Based on the strategy criteria established in (Wang 2007) and (Werbin 2010), seven variables were selected. In agreement with bank theory, four “natural” clusters were obtained, named “Personal”, “Commercial”, “Typical” and “Other banks”. In order to understand this grouping, projection pursuit based robust principal component analysis was conducted on the whole set showing that essentially three variables can be attributed the formation of different clusters. In order to reveal each group inner structure, we used R package mclust to fit a finite Gaussian mixture to the data. This revealed approximately a similar component structure, granting a common principal components analysis as in (Boente and Rodrigues, 2002). This allowed us to identify three variables which suffice for grouping and characterizing each cluster. Boente’s influence measures were used to detect extreme cases in the common principal components analysis. |
| format | Online |
| id | oai:ojs.revistas.unc.edu.ar:article-29385 |
| institution | Universidad Nacional de Cordoba |
| language | eng |
| publishDate | 2018 |
| publisher | Instituto de Economía y Finanzas. Facultad de Ciencias Económicas. Universidada Nacional de Córdoba. |
| record_format | ojs |
| spelling | oai:ojs.revistas.unc.edu.ar:article-293852022-04-13T04:20:20Z Robust Clustering of Banks in Argentina Agrupación robusta de Bancos en Argentina Vargas, José M. Díaz, Margarita García, Fernando Vargas, José M. Díaz, Margarita García, Fernando robust clustering projection pursuit common principal components robust K-means influence measures theory of the firm C23 G21 L2 agrupación robusta búsqueda de proyecciones componentes principales comunes K-media robusta medidas de influencia teoría de la empresa C23 G21 L2 The purpose of this paper is to classify and characterize 64 banks, active as of 2010 in Argentina, by means of robust techniques used on information gathered during the period 2001-2010. Based on the strategy criteria established in (Wang 2007) and (Werbin 2010), seven variables were selected. In agreement with bank theory, four “natural” clusters were obtained, named “Personal”, “Commercial”, “Typical” and “Other banks”. In order to understand this grouping, projection pursuit based robust principal component analysis was conducted on the whole set showing that essentially three variables can be attributed the formation of different clusters. In order to reveal each group inner structure, we used R package mclust to fit a finite Gaussian mixture to the data. This revealed approximately a similar component structure, granting a common principal components analysis as in (Boente and Rodrigues, 2002). This allowed us to identify three variables which suffice for grouping and characterizing each cluster. Boente’s influence measures were used to detect extreme cases in the common principal components analysis. El propósito de este documento es clasificar y caracterizar 64 bancos, activos en 2010 en la Argentina, mediante técnicas robustas utilizadas con información para el período 2001-2010. En base a los criterios de estrategia establecidos en (Wang 2007) y (Werbin 2010), se seleccionaron siete variables. De acuerdo con la teoría bancaria, se obtuvieron cuatro conglomerados "naturales", denominados "Personal", "Comercial", "Típico" y "Otros bancos". Para comprender este agrupamiento, se utilizó el todo el conjunto de banco y se realizó un análisis de los componentes principales basado en la proyección, que mostró que esencialmente tres variables pueden atribuirse a la formación de diferentes agrupaciones. A fin de revelar la estructura interna de cada grupo, utilizamos el paquete R mclust para ajustar una mezcla gaussiana finita a los datos. Esto reveló aproximadamente una estructura de componentes similar, lo que garantiza un análisis de componentes principales comunes como en (Boente y Rodrigues, 2002). Esto nos permitió identificar tres variables que son suficientes para agrupar y caracterizar cada cluster. Las medidas de influencia de Boente se utilizaron para detectar casos extremos en el análisis de componentes principales comunes. Instituto de Economía y Finanzas. Facultad de Ciencias Económicas. Universidada Nacional de Córdoba. 2018-12-01 info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion application/pdf https://revistas.unc.edu.ar/index.php/REyE/article/view/29385 10.55444/2451.7321.2018.v56.n1.29385 Revista de Economía y Estadística; Vol. 56 No. 1 (2018); 21-41 Revista de Economía y Estadística; Vol. 56 Núm. 1 (2018); 21-41 2451-7321 0034-8066 10.55444/2451.7321.2018.v56.n1 eng https://revistas.unc.edu.ar/index.php/REyE/article/view/29385/30185 Derechos de autor 2018 José M. Vargas, Margarita Díaz, Fernando García http://creativecommons.org/licenses/by-nc-nd/4.0 |
| spellingShingle | robust clustering projection pursuit common principal components robust K-means influence measures theory of the firm C23 G21 L2 agrupación robusta búsqueda de proyecciones componentes principales comunes K-media robusta medidas de influencia teoría de la empresa C23 G21 L2 Vargas, José M. Díaz, Margarita García, Fernando Vargas, José M. Díaz, Margarita García, Fernando Robust Clustering of Banks in Argentina |
| title | Robust Clustering of Banks in Argentina |
| title_alt | Agrupación robusta de Bancos en Argentina |
| title_full | Robust Clustering of Banks in Argentina |
| title_fullStr | Robust Clustering of Banks in Argentina |
| title_full_unstemmed | Robust Clustering of Banks in Argentina |
| title_short | Robust Clustering of Banks in Argentina |
| title_sort | robust clustering of banks in argentina |
| topic | robust clustering projection pursuit common principal components robust K-means influence measures theory of the firm C23 G21 L2 agrupación robusta búsqueda de proyecciones componentes principales comunes K-media robusta medidas de influencia teoría de la empresa C23 G21 L2 |
| topic_facet | robust clustering projection pursuit common principal components robust K-means influence measures theory of the firm C23 G21 L2 agrupación robusta búsqueda de proyecciones componentes principales comunes K-media robusta medidas de influencia teoría de la empresa C23 G21 L2 |
| url | https://revistas.unc.edu.ar/index.php/REyE/article/view/29385 |
| work_keys_str_mv | AT vargasjosem robustclusteringofbanksinargentina AT diazmargarita robustclusteringofbanksinargentina AT garciafernando robustclusteringofbanksinargentina AT vargasjosem robustclusteringofbanksinargentina AT diazmargarita robustclusteringofbanksinargentina AT garciafernando robustclusteringofbanksinargentina AT vargasjosem agrupacionrobustadebancosenargentina AT diazmargarita agrupacionrobustadebancosenargentina AT garciafernando agrupacionrobustadebancosenargentina AT vargasjosem agrupacionrobustadebancosenargentina AT diazmargarita agrupacionrobustadebancosenargentina AT garciafernando agrupacionrobustadebancosenargentina |