Multi-Period Portfolio Selection with Drawdown Control

Nystrup Peter, Madsen Henrik, Stephen Boyd, Erik Lindström

Research output: Contribution to conferencePaper, not in proceedingpeer-review

Abstract

In this talk, model predictive control (MPC) is used to dynamically optimize an investment portfolio. The
predictive control is based on multi-period forecasts of the mean and covariance of financial returns from a
multivariate hidden Markov model with time-varying parameters. Estimation and forecasting are done using an
online expectation--maximization algorithm. There are computational advantages to using MPC when estimates
of future returns are updated every time new observations become available, since the optimal control actions are
reconsidered anyway. Transaction and holding costs are important and are discussed as a means to address
estimation error and regularize the optimization problem. A complete practical implementation is presented
based on available market indices chosen to mimic the major liquid asset classes typically considered by an
institutional investor. In an out-of-sample test spanning two decades, the proposed approach to multi-period
portfolio selection successfully controls drawdowns with little or no sacrifice of mean--variance efficiency.
Original languageSwedish
Publication statusPublished - 2017 Jun 28
EventInternational Symposium on Forecasting, 2017 - Cairns, Australia
Duration: 2017 Jun 252017 Jun 28
Conference number: 37
https://isf.forecasters.org

Conference

ConferenceInternational Symposium on Forecasting, 2017
Country/TerritoryAustralia
CityCairns
Period2017/06/252017/06/28
Internet address

Subject classification (UKÄ)

  • Probability Theory and Statistics

Cite this