HISTORICAL PERFORMANCE: HYPOTHETICAL NATURE

The simple nature of the risk reduction models employed to generate the returns shown on the following pages should enable them to be effective portfolio management tools in the future. Moreover, the testing on historical data covers a large number of years and wide range of market conditions, both in secular bull and secular bear markets. The strong performance of the simple models over a long time and wide range of market conditions indicates that these models are quite likely to continue to be useful risk reduction tools.

It is important to note that the investment returns shown on the next pages are hypothetical. There are significant differences between hypothetical returns and returns from actual investments. In particular:

1. Hypothetical and backtested results do not result from actual trading using client assets. Although index mutual funds are the most likely investment vehicles and given their size it takes very large amounts of inflows or outflows to affect the broad markets, it is possible that MDP Associates trading of actual client accounts could have significantly affected the transaction prices.
2. The results shown may be from retroactive application of a model that was designed with the benefit of hindsight. (Since the models were not designed by MDP Associates, the extent to which this is true is unknown.)
3. Hypothetical performance information may not reflect the impact that any material market or economic factors might have had on MDP Associates' use of the investment methods if the methods had been used during the backtesting period in the actual management of client accounts

Perhaps the most important disclosure is the universal: past performance is no guarantee of future results. You should not assume that any of the returns shown here will be obtained in the future. It is possible that the methods and managed account clients will lose money.