Portfolio Resampling and Estimation Error

Bernd Scherer

This chapter introduces estimation error as an additional problem for traditional Markowitz-based portfolio construction. In contrast with Chapter 9, which also deals with estimation error but with a more decision-theoretic foundation, this chapter presents two heuristic methods that seem to be widespread among practitioners. In Sections 7.1 and 7.2 we use Monte Carlo methods to visualise the effect of estimation error on portfolio construction. We will distinguish between estimation error (when the distribution parameters are stable but we do not have enough data) and non-stationarity (the distribution parameters are unstable). The next four sections focus on the main topic of this chapter: resampled efficiency as a meaningful concept for portfolio construction.11The patent for the procedure, issued in December 1999 under the title “Portfolio Optimization by Means of Resampled Efficient Frontiers”, is held by its inventors, Richard and Robert Michaud. The procedure is worldwide patent pending and New Frontier Advisors, LLC, Boston, MA, has been granted exclusive worldwide licensing rights. The chapter concludes with an interpretation of investment constraints in the light of

To continue reading...

You need to sign in to use this feature. If you don’t have a Risk.net account, please register for a trial.

Sign in
You are currently on corporate access.

To use this feature you will need an individual account. If you have one already please sign in.

Sign in.

Alternatively you can request an individual account here: