Is there a point to conduct research to improve mean-variance optimization (MVO)? Because I understand that most of the poor performance in MVO is a result of the estimation error in expected returns.

Even in "Advanced Portfolio Management" by Paleologo, the author shows in Chapter 6.4 that a perfect forecast of volatility does not lead to substantial improvement in the Sharpe ratio when evaluating out-of-sample performance.

Main Question: Is research into other aspects of improving the MVO process fruitless unless we fix the estimation error issue w.r.t. expected returns?


1 Answer 1


Yes there is a big benefit in doing research on improving MVO. After all, the tangency portfolio is the best portfolio under several not super crazy assumptions.

There has been a lot of work and patents on MVO (the most common example is the re-sampling of the MV frontier). Both Michaud and Ibbotson patented different variations of the process:

a. Michaud’s patent - https://patents.google.com/patent/US6003018A/en (patent expired)

b. Ibbotson’s patent - https://patents.google.com/patent/US20030195831A1/en (patent abandoned after Michaud’s patent expired)

The patents have expired, but the filing documents describe the process in more detail as well if you do not have access to the book referenced above. Harry Markowitz and Nilufer Usmen published a paper in which they found that resampled mean variance optimization performed better than the original model in empirical tests, please refer to "Resampled Frontiers vs. Diffuse Bayes: An Experiment" Journal Of Investment Management Q4 2003" for more details.

  • 1
    $\begingroup$ The problem is fat tails and non normality of data, and intemporal inconsistency, a static equilibrium model is ineffective to capture real world dynamics, you have two choices, learn stochastic optimization models, or cvar risk optimization, or drawdown optimization, frankly mean variance Markowitz/Sharpe-Lintner modelling does not cut it in the real world, it is for professors because it is mathematically simple; and gullible MBA's who work for investment banks who do not question the reality of the status quo. There is a method developed by Black-Litterman you could try,which is effective. $\endgroup$ Commented Jun 13 at 21:38
  • $\begingroup$ Mean var optimisation models only work in bull markets, and are highly risk correlated and non-dynamic. $\endgroup$ Commented Jun 13 at 21:41
  • $\begingroup$ As they say when the market goes south and theres hell below , all shares go, the whole concept of mean variance diversification goes down the plug hole. $\endgroup$ Commented Jun 13 at 21:43
  • $\begingroup$ Thanks @phdstudent, I just finished the paper from JOIM and it was a very interesting read! This is the first time I have heard about resampling the efficient frontier and obtaining the best portfolio per a particular lambda. To your best knowledge, have these practices been employed in industry? Especially since resampling is so much less mathematically complex than the diffuse Bayes method. $\endgroup$
    – KaiSqDist
    Commented Jul 18 at 6:07
  • $\begingroup$ Michaud's fund uses it for sure. I use it on my own portfolio optimization too. $\endgroup$
    – phdstudent
    Commented Jul 18 at 18:57

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.