The Model Asset Allocations are provided by Ibbotson Associates, Inc., a registered investment advisor and wholly owed subsidiary of Morningstar, Inc. The Model Allocations are hypothetical in nature and contain generic asset classes that do not represent an actual portfolio, may not reflect the impact that material economic and market factors might have had on portfolio construction or an adviser’s decision-making if the adviser were actually managing clients’ assets, and do not reflect the deduction of advisory fees, brokerage or other commissions, and any other expenses that a client would have paid or actually paid.
In no way should the Model Allocations or their performance be considered indicative of or a guarantee of the future performance of a portfolio, nor should it be viewed as a substitute for the actual portfolios recommended to individual clients. Past performance does not guarantee future results. Results of an investment made today may differ substantially from historical performance.
When creating portfolios at the asset class level, investors should focus on two major qualifications: (1) efficiency from a mean-variance perspective, and (2) investor preferences. Portfolios that provide the best risk/return characteristics may not be acceptable to many clients due to counterintuitive allocations and investment biases. Furthermore, the most quantitatively efficient portfolios may not take into account possible errors in the input forecast. All of these factors ought to be incorporated into the portfolio recommendations.
Performing an unconstrained mean-variance optimization will often result in asset allocations that are not deemed practical by the investor and the investment professional. In an ideal world, the inputs used in mean-variance analysis would perfectly reflect future asset class behavior and would result in efficient portfolios that also meet investor “tolerances” for asset holdings. Unfortunately, this is not the case. Regardless of one’s method for calculating mean-variance analysis inputs, there will be instances where the resulting values differ dramatically from more qualitative expectations and investor tolerances. In addition, short-lived data series may result in unstable inputs. In both of these cases, it is necessary to constrain the allocations to such asset classes to reflect qualitative information, investor tolerances, or the lower confidence in various asset class inputs.