Optimal Portfolio Diversification via Independent Component Analysis

Título: Optimal Portfolio Diversification via Independent Component Analysis

Ponente: Victor De Miguel, Professor at LondonBusiness School

Organizador: Juan Francisco Monge Ivars

Fecha:  Lunes 4 de mayo a las 11:00 horas.

Lugar:  Online (Se os facilitara aquí el link el mismo día 30 minutos antes de la charla para que podáis acceder)

https://meet.google.com/neb-gdcm-ynx

Resumen: A popular approach to enhance portfoliodiversification is to use the factor-risk-parity portfolio, which is theportfolio whose return variance is spread equally among the principal components(PCs) of asset returns. Although PCs are unique and useful for dimensionreduction, they are an arbitrary choice because any rotation of the PCs remainsuncorrelated. This is problematic because we demonstrate that any portfolio isthe factor-variance-parity portfolio corresponding to a specific rotation ofthe PCs. To overcome this problem, we rely on the factor-risk-parity portfoliobased on the independent components (ICs), which are the rotation of the PCsthat are maximally independent, and thus, account for higher moments in assetreturns. We demonstrate that using the IC-variance-parity portfolio helps toreduce the return kurtosis. We also show how to exploit the near independenceof the ICs to parsimoniously estimate the factor-risk-parity portfolio based onValue-at-Risk. Finally, we empirically demonstrate that portfolios based on ICsoutperform those based on PCs, and the minimum-variance portfolio