This paper analyzes whether dynamically adjusting a portfolio with multiple asset classes can lead to superior returns. This paper utilizes mean reverting behavior of different asset classes and applies a relative valuation technique to dynamically allocate funds to six different asset classes. The dynamic asset allocation (DAA) strategy generates a positive annualized geometric mean return differential over 20 -30 year horizons. The standard deviation of the DAA strategy was lower than that of the six individual asset classes. The approach developed in this study can be useful to investors in identifying the most undervalued asset class at a particular point in time. This strategy may also be very well suited for retirement portfolios which are inherently long term in nature.