Sparse mean-reverting portfolios via penalized likelihood optimization. (January 2020)

• Record Type: Journal Article
• Authors: Zhang, Jize; Leung, Tim; Aravkin, Aleksandr
• Is Part Of: Automatica. Volume 111(2020)
• Publication Date: 2020-01
• Subjects: Automatic control -- Periodicals; Automation -- Periodicals; 629.805
• DOI: 10.1016/j.automatica.2019.108651
• Languages: English
• ISSNs: 0005-1098

Abstract

Abstract: An optimization approach is proposed to construct sparse portfolios with mean-reverting price behaviors. Our objectives are threefold: (i) design a multi-asset long-short portfolio that best fits an Ornstein–Uhlenbeck process in terms of maximum likelihood, (ii) select portfolios with desirable characteristics of high mean reversion through penalization, and (iii) select a parsimonious portfolio using ℓ 0 -regularization, i.e. find a small subset of a larger universe of assets that can be used for long and short positions. We present the full problem formulation, and develop a provably convergent algorithm for the nonsmooth, nonconvex objective based on partial minimization and projection. We demonstrate model functionalities on simulated and empirical price data, and include comparison with a pairs trading algorithm.