While IPEX can help clients to clarify their investment objectives, that is only half of the battle. IPEX must then analyze the overall investment market and determine the specific variables to be used in conducting the AAA. The starting point is the list of asset classes and investment styles that the client and IPEX have agreed will be considered as possible candidates for the new investment program. A four step process follows. 

First, select maximum allocation parameters. To ensure that the AAA produces a well diversified investment portfolio, that is not overly skewed in favor of certain asset classes or investment styles due to performance anomalies, IPEX will insert maximum allocation parameters for all of the asset classes and investment styles in the AAA. 

Second, establish expected returns. IPEX will project and calculate rates of return for all of the asset classes and investment styles to be included in the AAA. This step is arguably the most difficult, as it is essentially an exercise in trying to forecast the future. Even though the markets have produced prolonged periods of both above average and below average returns, these types of deviations can be misleading. Therefore, we normally rely upon long-term historical returns in making our estimates. 

Third, determine realistic levels of risk. An AAA should not provide the client with a false sense of security. It is critical that the client fully appreciate the level of risk associated with all of the asset classes and investment styles included in the analysis. Therefore, IPEX will analyze long-term data to determine realistic worst case scenarios and levels of volatility. Simply because an asset class or investment style has not produced an extremely negative return for a substantial number of years, that is no reason to believe that it cannot do so again in the near future.

Fourth, vary the assumptions. As historical averages can differ significantly depending upon the time period selected, IPEX will often choose multiple time periods, and prepare scenarios using different probabilities for expected returns and levels of volatility, to test the importance of these assumptions. This approach can be extremely helpful, as it demonstrates the importance of the underlying data, and dispels any notion that an AAA is a science that produces precise results.