Clustering Dynamics and Persistence for Financial Multivariate Panel Data

Tinbergen Institute Discussion Paper 2021-040/III

35 Pages Posted: 11 May 2021

See all articles by Igor Custodio João

Igor Custodio João

VU University Amsterdam

Andre Lucas

Vrije Universiteit Amsterdam - School of Business and Economics; Tinbergen Institute

Julia Schaumburg

Tinbergen Institute; VU University Amsterdam

Date Written: May 6, 2021

Abstract

We introduce a new method for dynamic clustering of panel data with dynamics for cluster location and shape, cluster composition, and for the number of clusters. Whereas current techniques typically result in (economically) too many switches, our method results in economically more meaningful dynamic clustering patterns. It does so by extending standard cross-sectional clustering techniques using shrinkage towards previous cluster means. In this way, the different cross-sections in the panel are tied together, substantially reducing short-lived switches of units between clusters (flickering) and the birth and death of incidental, economically less meaningful clusters. In a Monte Carlo simulation, we study how to set the penalty parameter in a data-driven way. A systemic risk surveillance example for business model classification in the global insurance industry illustrates how the new method works empirically.

Keywords: dynamic clustering, shrinkage, cluster membership persistence, Silhouette index, insurance

JEL Classification: G22, C33, C38

Suggested Citation

Custodio João, Igor and Lucas, Andre and Schaumburg, Julia and Schaumburg, Julia, Clustering Dynamics and Persistence for Financial Multivariate Panel Data (May 6, 2021). Tinbergen Institute Discussion Paper 2021-040/III, Available at SSRN: https://ssrn.com/abstract=3842954 or http://dx.doi.org/10.2139/ssrn.3842954

Igor Custodio João (Contact Author)

VU University Amsterdam ( email )

De Boelelaan 1105
Amsterdam, ND North Holland 1081 HV
Netherlands

Andre Lucas

Vrije Universiteit Amsterdam - School of Business and Economics ( email )

De Boelelaan 1105
Amsterdam, 1081 HV
Netherlands
+31 20 598 6039 (Phone)
+31 20 598 6020 (Fax)

HOME PAGE: http://personal.vu.nl/a.lucas

Tinbergen Institute

Roetersstraat 31
Amsterdam, 1018 WB
Netherlands

HOME PAGE: http://www.tinbergen.nl

Julia Schaumburg

VU University Amsterdam ( email )

De Boelelaan 1105
Amsterdam, ND North Holland 1081 HV
Netherlands
+31 (0)20 59 82653 (Phone)

HOME PAGE: http://juliaschaumburg.com

Tinbergen Institute ( email )

Gustav Mahlerplein 117
Amsterdam, 1082 MS
Netherlands

HOME PAGE: http://juliaschaumburg.com

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
24
Abstract Views
177
PlumX Metrics