The Black-Litterman Model and Views from a Reverse Optimization Procedure: An Out-of-Sample Performance Evaluation

The Black-Litterman (BL) model has been proposed as a valid solution to the problem of the estimation error in the parameter estimates of the Mean-Variance (MV) model. However, very little research has been done in order to facilitate the application of the BL model and to empirically test the performance of the BL model. The paper contributes to the existing literature by proposing an intuitive and easy application of the BL model. To this purpose, we suggest a novel approach to specify the investors views in the BL model. These views are derived by using a reverse optimization procedure similar to that used by the BL model in deriving the implied equilibrium expected excess returns. The second issue is addressed by empirically examining the out-of-sample performance of the BL model with respect to other asset allocation strategies. These strategies are given by the Sample MV (SMV), Minimum-Variance (MinV), Naive, Risk-Parity (RP), Capital Asset Pricing Model (CAPM), Strategic and the Bayesian strategy. In our empirical analysis, we also consider the reformulated BL model proposed by Allaj (2013) (BLEA model). Overall our results suggest that the BL model is a valid asset allocation strategy.