On the Ordering of Dynamic Principal Components and the Implications for Portfolio Analysis

33 Pages Posted: 17 May 2022

See all articles by Giovanni Bonaccolto

Giovanni Bonaccolto

Kore University of Enna - School of Economics and Law

Massimiliano Caporin

University of Padua - Department of Statistical Sciences

Date Written: April 26, 2022

Abstract

When principal component analysis (PCA) is used on a rolling or conditional setting, ordering and incoherence issues may emerge. We provide empirical evidence supporting this claim and introduce an algorithm that allows dynamic re-ordering of the principal components (PCs). We provide additional results that shed light on the consequences of incoherence when analyzing the link between PCs and macroeconomic risk factors, with a focus on the COVID-19 pandemic period. When PCs are optimally re-ordered, the role of factors emerges more clearly, with relevant implications for risk management.

Keywords: principal component analysis, dynamic principal component GARCH, risk factors, principal components ordering

JEL Classification: C38, C32, C58, G11, G17

Suggested Citation

Bonaccolto, Giovanni and Caporin, Massimiliano, On the Ordering of Dynamic Principal Components and the Implications for Portfolio Analysis (April 26, 2022). Available at SSRN: https://ssrn.com/abstract=4094084 or http://dx.doi.org/10.2139/ssrn.4094084

Giovanni Bonaccolto (Contact Author)

Kore University of Enna - School of Economics and Law ( email )

Italy

Massimiliano Caporin

University of Padua - Department of Statistical Sciences ( email )

Via Battisti, 241
Padova, 35121
Italy

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
99
Abstract Views
444
Rank
482,743
PlumX Metrics