Dynamic Asset Allocation Strategy Based on Estimated Excess Demand

This paper used a discrete time microstructure model studying the hidden excess demand and market liquidity of financial market which are two unobservable state variables. Based on the model, the estimates of the two immeasurable state variables may be obtained using the Kalman filter and the maximum likelihood method. Contrast to conventional financial model, we apply the excess demand to determine whether the market is overvalued or undervalued. A simple trading strategy for dynamic asset allocation, based on the estimated excess demand instead of the prediction for price, is proposed. Case studies on Cheung Kong from Hong Kong stock market show the proposed modeling and allocation strategy provide satisfactory control performance.