Classification can be useful in giving a synthetic and informative description of contextscharacterized by high degrees of complexity. Different approaches could be adopted to tacklethe classification problem: statistical tools may contribute to increase the degree of confidencein the classification scheme. A classi#cation algorithm for mutual funds style analysis is proposed,which combines di3erent statistical techniques and exploits information readily availableat low cost. Objective, representative, consistent and empirically testable classification schemesare strongly sought for in this field in order to give reliable information to investors and fundmanagers who are interested in evaluating and comparing di3erent #nancial products. Institutionalclassi#cation schemes, when available, do not always provide consistent and representative peergroups of funds. A “return-based” classi#cation scheme is proposed, which aims at identifyingmutual funds’ styles by analysing time series of past returns. The proposed classificationprocedure consists of three basic steps: (a) a dimensionality reduction step based on principalcomponent analysis, (b) a clustering step that exploits a robust evolutionary clustering methodology,and (c) a style identification step via a constrained regression model first proposed byWilliam Sharpe. The algorithm is tested on a sample of Italian mutual funds and achieves satisfactoryresults with respect to (i) the agreement with the existing institutional classi#cation and(ii) the explanatory power of out of sample variability in the cross-section of returns.

Clustering financial time series: an application tomutual funds style analysis / Pattarin, Francesco; Paterlini, Sandra; Minerva, Tommaso. - In: COMPUTATIONAL STATISTICS & DATA ANALYSIS. - ISSN 0167-9473. - STAMPA. - 47:2(2004), pp. 353-372. [10.1016/j.csda.2003.11.009]

Clustering financial time series: an application tomutual funds style analysis

PATTARIN, Francesco;PATERLINI, Sandra;MINERVA, Tommaso
2004

Abstract

Classification can be useful in giving a synthetic and informative description of contextscharacterized by high degrees of complexity. Different approaches could be adopted to tacklethe classification problem: statistical tools may contribute to increase the degree of confidencein the classification scheme. A classi#cation algorithm for mutual funds style analysis is proposed,which combines di3erent statistical techniques and exploits information readily availableat low cost. Objective, representative, consistent and empirically testable classification schemesare strongly sought for in this field in order to give reliable information to investors and fundmanagers who are interested in evaluating and comparing di3erent #nancial products. Institutionalclassi#cation schemes, when available, do not always provide consistent and representative peergroups of funds. A “return-based” classi#cation scheme is proposed, which aims at identifyingmutual funds’ styles by analysing time series of past returns. The proposed classificationprocedure consists of three basic steps: (a) a dimensionality reduction step based on principalcomponent analysis, (b) a clustering step that exploits a robust evolutionary clustering methodology,and (c) a style identification step via a constrained regression model first proposed byWilliam Sharpe. The algorithm is tested on a sample of Italian mutual funds and achieves satisfactoryresults with respect to (i) the agreement with the existing institutional classi#cation and(ii) the explanatory power of out of sample variability in the cross-section of returns.
2004
47
2
353
372
Clustering financial time series: an application tomutual funds style analysis / Pattarin, Francesco; Paterlini, Sandra; Minerva, Tommaso. - In: COMPUTATIONAL STATISTICS & DATA ANALYSIS. - ISSN 0167-9473. - STAMPA. - 47:2(2004), pp. 353-372. [10.1016/j.csda.2003.11.009]
Pattarin, Francesco; Paterlini, Sandra; Minerva, Tommaso
File in questo prodotto:
File Dimensione Formato  
Pattarin, Paterlini & Minerva (2004).pdf

Accesso riservato

Dimensione 322.35 kB
Formato Unknown
322.35 kB Unknown   Visualizza/Apri   Richiedi una copia
Pubblicazioni consigliate

Licenza Creative Commons
I metadati presenti in IRIS UNIMORE sono rilasciati con licenza Creative Commons CC0 1.0 Universal, mentre i file delle pubblicazioni sono rilasciati con licenza Attribuzione 4.0 Internazionale (CC BY 4.0), salvo diversa indicazione.
In caso di violazione di copyright, contattare Supporto Iris

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11380/12748
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 65
  • ???jsp.display-item.citation.isi??? 52
social impact