A General Comovement Measure for Time Series

Research output: Chapter in Book/Report/Conference proceedingChapterScientificpeer-review


We propose a nonparametric, time-dependent, cross-scale/cross-frequency dependence measure for multivariate stationary and non-stationary time series termed multi-thickness thick pen measure of association, MTTPMA. The building blocks of the measure are the Thick Pen Transform and the Thick Pen Measure of Association. The new measure is simple and visually interpretable. We demonstrate its potential application on synthetic financial contagion data.
Original languageEnglish
Title of host publicationMathematical and Statistical Methods for Actuarial Sciences and Finance : eMAF2020
EditorsMarco Corazza, Manfred Gilli, Cira Perna, Claudio Pizzi, Marilena Sibillo
Place of PublicationCham
Publication date2021
ISBN (Print)978-3-030-78964-0, 978-3-030-78967-1
ISBN (Electronic)978-3-030-78965-7
Publication statusPublished - 2021
MoE publication typeA3 Book chapter

Publication series

NameMathematical and Statistical Methods for Actuarial Sciences and Finance


  • 112 Statistics and probability
  • cross-scale
  • multiscale
  • multivariate
  • nonparametric
  • time-varying


Dive into the research topics of 'A General Comovement Measure for Time Series'. Together they form a unique fingerprint.

Cite this