Dynamic clustering of multivariate panel data

We propose a dynamic clustering model for uncovering latent time-varying group structures in multivariate panel data. The model is dynamic in three ways. First, the cluster location and scale matrices are time-varying to track gradual changes in cluster characteristics over time. Second, all units c...

Full description

Saved in:
Bibliographic Details
Published in:Policy File 2021
Main Authors: Igor Custodio Joao, Lucas, André, Schaumburg, Julia, Schwaab, Bernd
Format: Report
Language:English
Subjects:
Banking
Economic models
Interest rates
Markov analysis
Multivariate analysis
Profitability
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites
container_end_page
container_issue
container_start_page
container_title Policy File
container_volume
creator Igor Custodio Joao
Lucas, André
Schaumburg, Julia
Schwaab, Bernd
description We propose a dynamic clustering model for uncovering latent time-varying group structures in multivariate panel data. The model is dynamic in three ways. First, the cluster location and scale matrices are time-varying to track gradual changes in cluster characteristics over time. Second, all units can transition between clusters based on a Hidden Markov model (HMM). Finally, the HMM’s transition matrix can depend on lagged time-varying cluster distances as well as economic covariates. Monte Carlo experiments suggest that the units can be classified reliably in a variety of challenging settings. Incorporating dynamics in the cluster composition proves empirically important in an a study of 299 European banks between 2008Q1 and 2018Q2. We find that approximately 3% of banks transition per quarter on average. Transition probabilities are in part explained by differences in bank profitability, suggesting that low interest rates can lead to long-lasting changes in financial industry structure.
format report
fullrecord <record><control><sourceid>proquest_AOXKD</sourceid><recordid>TN_cdi_proquest_reports_3153121014</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>3153121014</sourcerecordid><originalsourceid>FETCH-proquest_reports_31531210143</originalsourceid><addsrcrecordid>eNrjZNB1qcxLzM1MVkjOKS0uSS3KzEtXyE9TyC3NKcksSyzKTCxJVShIzEvNUUhJLEnkYWBNS8wpTuWF0twMSm6uIc4eugVF-YWlqcUl8UWpBflFJcXxxoamxoZGhgaGJsZEKQIAYgYsIw</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>report</recordtype><pqid>3153121014</pqid></control><display><type>report</type><title>Dynamic clustering of multivariate panel data</title><source>Policy File Index</source><creator>Igor Custodio Joao ; Lucas, André ; Schaumburg, Julia ; Schwaab, Bernd</creator><creatorcontrib>Igor Custodio Joao ; Lucas, André ; Schaumburg, Julia ; Schwaab, Bernd</creatorcontrib><description>We propose a dynamic clustering model for uncovering latent time-varying group structures in multivariate panel data. The model is dynamic in three ways. First, the cluster location and scale matrices are time-varying to track gradual changes in cluster characteristics over time. Second, all units can transition between clusters based on a Hidden Markov model (HMM). Finally, the HMM’s transition matrix can depend on lagged time-varying cluster distances as well as economic covariates. Monte Carlo experiments suggest that the units can be classified reliably in a variety of challenging settings. Incorporating dynamics in the cluster composition proves empirically important in an a study of 299 European banks between 2008Q1 and 2018Q2. We find that approximately 3% of banks transition per quarter on average. Transition probabilities are in part explained by differences in bank profitability, suggesting that low interest rates can lead to long-lasting changes in financial industry structure.</description><language>eng</language><publisher>European Central Bank</publisher><subject>Banking ; Economic models ; Interest rates ; Markov analysis ; Multivariate analysis ; Profitability</subject><ispartof>Policy File, 2021</ispartof><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://www.proquest.com/docview/3153121014?pq-origsite=primo$$EHTML$$P50$$Gproquest$$H</linktohtml><link.rule.ids>776,780,4476,43724,72839,72844</link.rule.ids><linktorsrc>$$Uhttps://www.proquest.com/docview/3153121014?pq-origsite=primo$$EView_record_in_ProQuest$$FView_record_in_$$GProQuest</linktorsrc></links><search><creatorcontrib>Igor Custodio Joao</creatorcontrib><creatorcontrib>Lucas, André</creatorcontrib><creatorcontrib>Schaumburg, Julia</creatorcontrib><creatorcontrib>Schwaab, Bernd</creatorcontrib><title>Dynamic clustering of multivariate panel data</title><title>Policy File</title><description>We propose a dynamic clustering model for uncovering latent time-varying group structures in multivariate panel data. The model is dynamic in three ways. First, the cluster location and scale matrices are time-varying to track gradual changes in cluster characteristics over time. Second, all units can transition between clusters based on a Hidden Markov model (HMM). Finally, the HMM’s transition matrix can depend on lagged time-varying cluster distances as well as economic covariates. Monte Carlo experiments suggest that the units can be classified reliably in a variety of challenging settings. Incorporating dynamics in the cluster composition proves empirically important in an a study of 299 European banks between 2008Q1 and 2018Q2. We find that approximately 3% of banks transition per quarter on average. Transition probabilities are in part explained by differences in bank profitability, suggesting that low interest rates can lead to long-lasting changes in financial industry structure.</description><subject>Banking</subject><subject>Economic models</subject><subject>Interest rates</subject><subject>Markov analysis</subject><subject>Multivariate analysis</subject><subject>Profitability</subject><fulltext>true</fulltext><rsrctype>report</rsrctype><creationdate>2021</creationdate><recordtype>report</recordtype><sourceid>ABWIU</sourceid><sourceid>AFVLS</sourceid><sourceid>ALSLI</sourceid><sourceid>AOXKD</sourceid><sourceid>DPSOV</sourceid><recordid>eNrjZNB1qcxLzM1MVkjOKS0uSS3KzEtXyE9TyC3NKcksSyzKTCxJVShIzEvNUUhJLEnkYWBNS8wpTuWF0twMSm6uIc4eugVF-YWlqcUl8UWpBflFJcXxxoamxoZGhgaGJsZEKQIAYgYsIw</recordid><startdate>20210729</startdate><enddate>20210729</enddate><creator>Igor Custodio Joao</creator><creator>Lucas, André</creator><creator>Schaumburg, Julia</creator><creator>Schwaab, Bernd</creator><general>European Central Bank</general><scope>ABWIU</scope><scope>AFVLS</scope><scope>ALSLI</scope><scope>AOXKD</scope><scope>DPSOV</scope></search><sort><creationdate>20210729</creationdate><title>Dynamic clustering of multivariate panel data</title><author>Igor Custodio Joao ; Lucas, André ; Schaumburg, Julia ; Schwaab, Bernd</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_reports_31531210143</frbrgroupid><rsrctype>reports</rsrctype><prefilter>reports</prefilter><language>eng</language><creationdate>2021</creationdate><topic>Banking</topic><topic>Economic models</topic><topic>Interest rates</topic><topic>Markov analysis</topic><topic>Multivariate analysis</topic><topic>Profitability</topic><toplevel>online_resources</toplevel><creatorcontrib>Igor Custodio Joao</creatorcontrib><creatorcontrib>Lucas, André</creatorcontrib><creatorcontrib>Schaumburg, Julia</creatorcontrib><creatorcontrib>Schwaab, Bernd</creatorcontrib><collection>Social Science Premium Collection</collection><collection>Policy File Index</collection><collection>Politics Collection</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Igor Custodio Joao</au><au>Lucas, André</au><au>Schaumburg, Julia</au><au>Schwaab, Bernd</au><format>book</format><genre>unknown</genre><ristype>RPRT</ristype><atitle>Dynamic clustering of multivariate panel data</atitle><jtitle>Policy File</jtitle><date>2021-07-29</date><risdate>2021</risdate><abstract>We propose a dynamic clustering model for uncovering latent time-varying group structures in multivariate panel data. The model is dynamic in three ways. First, the cluster location and scale matrices are time-varying to track gradual changes in cluster characteristics over time. Second, all units can transition between clusters based on a Hidden Markov model (HMM). Finally, the HMM’s transition matrix can depend on lagged time-varying cluster distances as well as economic covariates. Monte Carlo experiments suggest that the units can be classified reliably in a variety of challenging settings. Incorporating dynamics in the cluster composition proves empirically important in an a study of 299 European banks between 2008Q1 and 2018Q2. We find that approximately 3% of banks transition per quarter on average. Transition probabilities are in part explained by differences in bank profitability, suggesting that low interest rates can lead to long-lasting changes in financial industry structure.</abstract><pub>European Central Bank</pub></addata></record>
fulltext fulltext_linktorsrc
identifier
ispartof Policy File, 2021
issn
language eng
recordid cdi_proquest_reports_3153121014
source Policy File Index
subjects Banking
Economic models
Interest rates
Markov analysis
Multivariate analysis
Profitability
title Dynamic clustering of multivariate panel data
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-10T16%3A58%3A42IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_AOXKD&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=unknown&rft.atitle=Dynamic%20clustering%20of%20multivariate%20panel%20data&rft.jtitle=Policy%20File&rft.au=Igor%20Custodio%20Joao&rft.date=2021-07-29&rft_id=info:doi/&rft_dat=%3Cproquest_AOXKD%3E3153121014%3C/proquest_AOXKD%3E%3Cgrp_id%3Ecdi_FETCH-proquest_reports_31531210143%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=3153121014&rft_id=info:pmid/&rfr_iscdi=true