We are very proud to introduce AllocationADVISOR 6.0. It represents over two years of work driven by our desire to provide you with a tool for realizing the benefits of mean-variance optimization.

New Features in AllocationADVISOR 6.0:

• New Allocation Case Dialog
• Black-Litterman Return Forecasts
• Multiple Allocation Cases in one workbook
• Display two frontiers on the Efficient Frontier Graph
• Multiple Custom Portfolios
• Moving Average Flows in Monte Carlo Simulation
• New colors on the Simulation Probabilities Graph

New Allocation Case Window and Multiple Allocation Cases

AllocationADVISOR 6.0 has a dramatically improved design for entering the inputs for allocation analysis. Much like the Monte Carlo Simulation Cases, AllocationADVISOR now allows you to build multiple Allocation Cases in one workbook. The ability to create multiple Asset Allocation cases makes it very easy to compare changes in inputs and assumptions. You can even view two frontiers on the same graph.

The inputs for Allocation Cases are entered in the Allocation Case “tree” which appears in the Allocation Case window. This tree puts the inputs for multiple cases at your fingertips. Experienced AllocationADVISOR users will find the pages in the Allocation Case window very familiar, as many of them are based on the Analysis Definition window.

What about my current AllocationADVISOR files?

All of your existing AllocationADVISOR 5.0 files are fully compatible with AllocationADVISOR 6.0. The Historical Forecast Model is the same as the previous AllocationADVISOR forecast model; it uses historical estimates of returns, standard deviations, and correlations to create an efficient frontier. However, in AllocationADVISOR 6.0 the controls from the old Analysis Definition dialog are divided into separate windows, making it easier to manage your analysis.

Getting Started

Now that you can have multiple Asset Allocation cases in a single workbook or file, it is important to name your cases. To create an Asset Allocation case, enter a name for the case and select which Forecast Model you would like to use. AllocationADVISOR 6.0 includes two Forecast Models – Historical and Black-Litterman. The Historical model uses historical data to create the forecasts and is identical to the model used in previous versions of AllocationADVISOR. The Black-Litterman model is described below.

What is the Black-Litterman model?

The biggest new feature of AllocationADVISOR 6.0 is the inclusion of the Black-Litterman model. The Black-Litterman model was created by Fischer Black and Robert Litterman of Goldman Sachs. Conceptually, it combines a number of the pillars of modern portfolio theory – Sharpe’s CAPM and Markowitz’s mean-variance optimization. The Black-Litterman model is not an alternative or a replacement of mean-variance optimization; it is a tool for creating a set of expected returns for use within the mean-variance optimization framework.

Harry Markowitz’s mean-variance optimization is the heart of modern asset allocation. Unfortunately, a number of people don’t use mean-variance optimization because the resulting asset allocations are highly concentrated in just a few of the assets being optimized. The Black-Litterman model eliminates this problem by creating better estimates of expected return that carefully balance risk and return, leading to well-diversified portfolios.

For more information about how the Black-Litterman Forecast Model leads to diversified portfolios, see the article Black-Litterman: Asset Allocations You Can Actually Use! For a discussion about how the Black-Litterman model calculate forecasts, see How does the Black-Litterman Model Calculate Return Forecasts?

How do you create a Black-Litterman Asset Allocation Case?

We have gone to great lengths to make this sophisticated model easy to use. To create an Asset Allocation case using the Black-Litterman Forecast model, after naming the Allocation Case and selecting the Black-Litterman Forecast Model, you begin by selecting the assets to optimize. Zephyr groups some of the most popular asset classes together in what we call Asset Palettes. Technically, Asset Palettes are collections of Market Cap Assets, where Market Cap Assets are asset class index proxies that are linked to an estimate of the market capitalization of the asset class in question. Using Asset Palettes created by Zephyr makes it very easy to select a logical set of asset classes. Zephyr-created Asset Palettes are groups of non-overlapping asset classes that represent reasonable market portfolios or segments of a market portfolio. AllocationADVISOR also allows users to create Custom Palettes. The ability to select a predefined Asset Palette can dramatically decrease the time required to create an asset allocation analysis.

After selecting the Asset Palette, you need to enter a Risk-Premium and a Risk-Free Rate. For those that need assistance in estimating these values we provide the historical 10, 25, and 50 year risk premia on six model portfolios, as well as the annual yield on the 3-month, 1-year, 5-year, and 10-year US Treasuries. The historical risk-premia and annual yield data is updated monthly and distributed with Zephyr’s Full Index Update.

After specifying a Palette Risk Premium and a Risk-Free Rate, all that you need to do is select Done and an Efficient Frontier graph, Portfolio table and Inputs table for the Allocation Case are added to your workbook.

How do you incorporate your unique forecasts of returns?

For those of you who wish to modify the return forecasts to match your own opinions about future market performance, the Black-Litterman model provides an elegant framework for combining a base case set of market implied returns with your unique forecasts. The new mixed Forecast Returns still lead to well-diversified portfolios that reflect your opinions.

Your unique forecasts of expected returns are called Views. There are two types of Views – Absolute Views and Relative Views – each of which is entered on their respective windows in the Allocation Case tree.

Absolute Views state your unique return forecast of an asset class or group of asset classes. For example, US Small Cap will have a return of 11.5%.

US Small Cap is actually a combination of two asset classes – US Small Cap Growth and US Small Cap Value. The formation of View Groups gives you greater flexibility for specifying views.

Relative Views specify the expected return differential between two asset classes / view groups. For example, US Large Cap Growth will outperform US Large Cap Value by 1%.

For both types of views, you need to specify a confidence level between 5% and 95%. This confidence level represents the strength of your view. All else equal, the more confidence you assign to a View, the more aggressively the model will implement that view.

Once you have specified your views, select Done.

How do you show two efficient frontiers on the same graph?

One of the most requested features that has been added to AllocationADVISOR is the ability to show two efficient frontiers on the same graph.

In addition to two efficient frontier graphs, you can also plot the benchmark on the Efficient Frontier graph. The efficient frontier options are available from a right-click menu.

Multiple Custom Portfolios

AllocationADVISOR 6.0 allows users to enter multiple custom portfolios in their allocation cases. These can be any portfolio that users wish to compare to the efficient portfolios, such as a current portfolio or a target portfolio. Users can select which of the custom portfolios will be the Comparison Portfolio, which can be displayed on many of the graphs and tables.

Monte Carlo Enhancements

We also made two enhancements to the Monte Carlo Simulations for this release. The first is the ability to enter moving average cash flows. We have also changed the colors on the Simulation Probability graph, making it easier to distinguish between the different probabilities.

Conclusion

AllocationADVISOR 6.0 includes the Black-Litterman model, one of the most sophisticated asset allocation models. Best of all, we have removed the complexity associated with the model making it intuitive and very easy to use. At last, you will be able use a mean-variance optimizer and feel good about the resulting allocations. A copy of the workbook used to create the sample illustrations is available for download:

New In AllocationADVISOR 6.0 (.exe file)

New In AllocationADVISOR 6.0 (.zip file)

For additional information on the Black-Litterman model, see the article “Black-Litterman: Asset allocation you can actually use!”

Back to the Newsletters page.