Optimizing a portfolio of mean-reverting assets with transaction costs via a feedforward neural network

John M. Mulvey, Yifan Sun, Mengdi Wang, Jing Ye

Research output: Contribution to journalArticlepeer-review

Abstract

Optimizing a portfolio of mean-reverting assets under transaction costs and a finite horizon is severely constrained by the curse of high dimensionality. To overcome the exponential barrier, we develop an efficient, scalable algorithm by employing a feedforward neural network. A novel concept is to apply HJB equations as an advanced start for the neural network. Empirical tests with several practical examples, including a portfolio of 48 correlated pair trades over 50 time steps, show the advantages of the approach in a high-dimensional setting. We conjecture that other financial optimization problems are amenable to similar approaches.

Original languageEnglish (US)
JournalQuantitative Finance
DOIs
StateAccepted/In press - Jan 1 2020

All Science Journal Classification (ASJC) codes

  • Finance
  • Economics, Econometrics and Finance(all)

Keywords

  • Asset allocation
  • Portfolio allocation
  • Portfolio optimization
  • Statistical learning theory
  • Stochastic programming

Fingerprint Dive into the research topics of 'Optimizing a portfolio of mean-reverting assets with transaction costs via a feedforward neural network'. Together they form a unique fingerprint.

Cite this