Resampled efficient frontier VS Mean-variance optimization

Schweser Notes 2 Page 240 Question 4 Just a question about asset allocation. Here is the reading: Robinson begins to select a static asset allocation for each investor given the information he gathered from both interviews. His first step entails specifying asset classes. Robinson believes that the set of asset classes should provide a high level of diversification and they should have a large percentage of liquid assets. However, he does not think that a majority of all possible investable assets need to be included in a given portfolio or that assets need to be classified into more than one class. A drawback to Robinson’s asset allocation approach is that the number of estimates needed is overwhelming. Robinson evaluates the following asset classes as possible investments for Carlson and Olsen: Asset Class Expected Return Expected Standard Deviation U.S. Large-Cap 8.5% 15% U.S. Small-Cap 12% 20% U.S. Fixed Income 5.5% 3% Real Estate 7.0% 12% And the question: What approach to asset allocation is Robinson most likely using? A. Resampled efficient frontier. B. Mean-variance optimization. C. Black-Litterman. The answer is B and mentioned:“Mean-variance optimization uses a static approach as opposed to a dynamic approach. Also, the primary drawback to mean-variance optimization is the overwhelming number of estimates needed (e.g., expected returns, standard deviations, correlations).” I do not quite understand this. For me, the overwhelming number of estimates also applies to “Resampled efficient frontier” and “Black-Litterman”

I chose B before scrolling down. If you remember the SD for a portfolio calculation we learned at L1 and L2, it can multiply in size rather quickly as you add more assets to the mix. I don’t know if this answers your question, but that was my thought process.

Thanks a lot for the answer. I understand the overwhelming calculation about the SD and agree with you that the answer B is correct. However, from my understanding, the calculation of SD also applies to Resampled efficient frontier. Therefore, answer A is also collect. Am I right?

lawbringer Wrote: ------------------------------------------------------- > Thanks a lot for the answer. I understand the > overwhelming calculation about the SD and agree > with you that the answer B is correct. However, > from my understanding, the calculation of SD also > applies to Resampled efficient frontier. > Therefore, answer A is also collect. Am I right? No, because the question asked is “approach to asset allocation is Robinson most likely using”. The simplicity in determining the asset allocation points to answer B.