A Dynamic Latent-Space Model for Asset Clustering

Volatility clustering Soar
DOI: 10.1515/snde-2022-0111 Publication Date: 2023-11-23T12:40:19Z
ABSTRACT
Abstract Periods of financial turmoil are not only characterized by higher correlation across assets but also modifications in their overall clustering structure. In this work, we develop a dynamic Latent-Space mixture model for capturing changes the structure at fine scale. Through model, able to project stocks onto lower dimensional manifold and detect presence clusters. The infinite-mixture assumption ensures tractability inference accommodates cases which number clusters is large. Bayesian framework rely on accounts uncertainty parameters’ space allows inclusion prior knowledge. After having tested our model’s effectiveness suitable synthetic dataset, apply cross-correlation series two reference stock indices. Our correctly captures time-varying asset clustering. Moreover, notice how assets’ latent coordinates may be related relevant factors such as market capitalization volatility. Finally, find further evidence that seems soar periods distress.
SUPPLEMENTAL MATERIAL
Coming soon ....
REFERENCES (62)
CITATIONS (1)
EXTERNAL LINKS
PlumX Metrics
RECOMMENDATIONS
FAIR ASSESSMENT
Coming soon ....
JUPYTER LAB
Coming soon ....