Factor Analysis of Moving Average Processes

TL;DR

This paper extends factor analysis to moving average processes by formulating it as a spectral density rank minimization problem and proposing a convex relaxation approach.

Contribution

It introduces a novel method for factor analysis of moving average processes using spectral density rank minimization and convex relaxation techniques.

Findings

Effective approximation of spectral density rank minimization

Convex relaxation via trace norm is suitable for the problem

Provides a new framework for factor analysis in time series

Abstract

The paper considers an extension of factor analysis to moving average processes. The problem is formulated as a rank minimization of a suitable spectral density. It is shown that it can be adequately approximated via a trace norm convex relaxation.