April 2021

Dynamic Ingredient Copula Devices sms lån på minuttet Estimated Cluster Assignments

Dong Hwan Oh and Andrew J. Patton


This paper proposes a dynamic multi-facet copula to be used in excessive dimensional time series functions. A recent feature of our mannequin is that the assignment of particular person variables to groups is estimated from the records, in set of being pre-assigned the utilize of SIC industry codes, market capitalization ranks, or other ad hoc systems. We adapt the okay-procedure clustering algorithm to be used in our utility and gift that it has very advantageous finite-pattern properties. Applying the contemporary mannequin to returns on 110 US equities, we acquire round 20 clusters to be optimum. In out-of-pattern forecasts, we acquire that a mannequin sms lån på minuttet as few as 5 estimated clusters very a lot outperforms an in another case identical mannequin sms lån på minuttet 21 clusters forsms lån på minuttet the utilize of two-digit SIC codes.

Keywords: correlation, tail risk, multivariate density forecast

DOI: https://doi.org/10.17016/FEDS.2021.029

Plump Paper

Assist to Top

Final Change:
April 30, 2021

Leave a Reply

Your email address will not be published. Required fields are marked *