The Quality of Blume and Vasicek Betas for forecasting systematic risk: Evidence from a German stock portfolio

Abstract In both practical applications and empirical capital market research the beta factor is often estimated for the future period using an OLS regression analysis based on historical data. However, these betas are often not reliable for the future. For this reason, methods have been developed to correct or adjust the (raw) beta. The aim of this study is to find out to what extent the forecasting quality of a stock beta can be improved by Blume Beta, Vasicek Beta and a simple variant of Blume Beta ("Adjusted Beta"). For this purpose, an empirical analysis is carried out on the basis of 10 stocks included in the German stock index DAX. For the entire period under review, it can be observed that all three adjustment methods tend to improve the forecast quality compared to the raw beta, with the simple “Adjusted beta” leading to the best values. The examination of two subperiods, one covering the financial and economic crisis of 2008/2009 and the other the Corona pandemic, shows that the forecasting quality tends to be lower when looking at individual stocks compared with the overall period. However, according to the values (i.e. averages of all stocks) of the forecasting quality measures used in this paper, the adjustment procedures mostly produce better values than raw beta in both subperiods. JEL classification numbers: G11. Keywords: Adjusted beta, Blume beta, Vasicek beta, Mean Squared Error.