Cointegration and asset allocation: a new active hedge fund strategy

March 2001 Abstract The main objective of a hedge strategy is to generate positive returns irrespective of market conditions. This paper presents a classic hedge fund strategy: an investment vehicle whose key objective is to mi nimize investment risk in an attempt to deliver profits under all market circumstances. The hedge is designed to have minimal correlation with the market and, irrespective of market direction, the fund seeks to generate positive alpha. A significant diffe rence between this model and more traditional hedge fund strategies is that portfolio optimization is based upon the cointegration of prices rather than the correlation of returns. Models that are based on mean -variance analysis seek portfolio weights to minimise the variance of the portfolio for a given level of return. The portfolio variance is measured using a covariance matrix and these matrices are notoriously difficult to estimate. Moreover the mean -variance criterion has nothing to ensure that trac king errors are stationary. Although the portfolios will be efficient, the tracking errors will in all probability be random walks. Therefore the replicating portfolio can drift very far from the benchmark unless it is frequently re -balanced. This paper sh ows that it is possible to devise allocations that have stationary tracking errors: any strategy that guarantees a stationary tracking error must be based on cointegration. Efficient long short hedge strategies may be achieved with relatively few stocks an d with much lower turnover rates compared to traditional market neutral strategies . JEL Classification : C32, G11, G15