Faktor Investeringsstrategier: Porteføljekonstruktion med Fama-French Risikofaktorer

Jonas Riisgaard Koefoed & Mads Skak Bossen

Student thesis: Master thesis

Abstract

This thesis makes an analysis into Fama & French methods on the US S&P 500 index and aims at uncovering whether these methods are appropriate to determine return pattern effects and to be used as investment strategies. The thesis is departing from a theoretical academic perspective, where the theory behind the Fama & French factor models is sub-stantiated. This analysis is taken even further by introducing new factors to the traditional three-, four- and five-factor model of Fama & French. The main analysis is divided into two parts. The first part seeks to examine the models’ ca-pability to explain/clarify variations in cross sections returns and thus be used as a pricing mechanism for assets. The result showed that adding new factors to the five-factor model, and thus creating a seven-factor model gave a significantly better result. The seven-factor model was able to explain 95% of the cross section variance and had a rejection rate of ex-plaining the dependent left hand side portfolios of 10%. Further analysis showed that this model could be reduced to a five-factor model with the same ability to explain cross section variance, but with a 2% higher rejection rate. The aim of the second part was to analyse whether the different factors of the Fama & French model can be utilised as investment strategies. While some of the individual fa-ctors showed potential, the combination of two or more factors was severely increasing the performance of the portfolios. Furthermore, it was analysed that dynamic factor-portfolios combining individual factors, based on the historical observation of return and Sharpe-Ratios, was the most effective strategy. Thus, the combination of several factors, diversi-fying risk and thereby increasing the probability of high risk-adjusted returns is argued to be preferable. Conclusive, this thesis has shown that Fama & French’s models are suitable for the examina-tion of cross section returns and to be used as investment strategies for firms on the S&P 500 index. Specifically, with regards to the risk of a recession, the diversification of the factors applied, spreading risk across the market, is compelling.

Educations, (Graduate Programme) Final Thesis
LanguageDanish
Publication date2020
Number of pages123
SupervisorsMarcel Fischer