The Black-Litterman Model : An Application on the Norwegian Stock Market

Abstract

In this thesis we apply the Black-Litterman model on the Norwegian stock market using historical price data in the period of 2004 to 2015. A wide collection of analyst recommendations was used to determine views to feed into the model. We provide a theoretical framework of the model, and discuss the implications of some of the approaches in the literature. To best understand the functioning of the model, we compare it to several mean-variance models and pure benchmark portfolios by evaluating them based on five criteria. They asses return, both risk adjusted and not, transaction cost and predictability. With equal weights on all criteria, the Black-Litterman portfolios perform mediocre despite a positive contribution from the views. Regardless of its ranking among comparable portfolios, the model behaves intuitively and is undoubtedly an upgrade to Markowitz traditional method.