Skip to content
All library documents

Building Sparse Mean-Reverting Portfolios with Penalized OU Likelihood

Article arXiv papers · Author: Jize Zhang et al.
Read the original

Summary

This paper presents an optimization method for constructing a mean-reverting portfolio from a larger set of assets. It seeks portfolios whose spread is well described by an Ornstein–Uhlenbeck process, with process parameters estimated by maximum likelihood. The stated selection goals combine stronger mean reversion and lower variance with parsimony, so that the resulting portfolio uses a smaller subset of assets for long and short positions.

The authors formulate the optimization problem and describe a specialized algorithm that uses partial minimization. They illustrate the method with numerical examples based on simulated and empirical price data. The abstract does not specify the penalty form, asset universe, trading rules, transaction costs, or out-of-sample performance. The examples therefore indicate how the estimation and selection approach is demonstrated, but the available description is insufficient to judge its stability or profitability in live trading.

Key ideas

  • The method selects portfolios intended to follow an Ornstein–Uhlenbeck process.
  • OU parameters are estimated with maximum likelihood.
  • Portfolio selection targets high mean reversion and low variance.
  • A parsimony objective encourages using fewer assets for long and short positions.
  • A specialized algorithm uses partial minimization, with examples on simulated and empirical prices.

Tags

Mean reversionPairs tradingPortfolio constructionStatistics
Full text
# Mean Reverting Portfolios via Penalized OU-Likelihood Estimation


# Mean Reverting Portfolios via Penalized OU-Likelihood Estimation









We study an optimization-based approach to con- struct a mean-reverting portfolio of assets. Our objectives are threefold: (1) design a portfolio that is well-represented by an Ornstein-Uhlenbeck process with parameters estimated by maximum likelihood, (2) select portfolios with desirable characteristics of high mean reversion and low variance, and (3) select a parsimonious portfolio, 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, a specialized algorithm that exploits partial minimization, and numerical examples using both simulated and empirical price data.

Shown in full with attribution under the source's licence. Licence: abstract CC0

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.