Value og momentum i kombination: En balanceret vej til merafkast

Thomas Ho Vestergaard Pedersen & Morten Brogaard Marthedal

Student thesis: Master thesis


A vast number of academic studies have documented ubiquitously abnormal returns to valueoriented as well as momentum-oriented investors. Using data, free of survivorship bias, we show further evidence of both a value and momentum premium on the Scandinavian stock market since 1991. To mitigate data mining concerns, we apply a single, simple and uniform value and momentum measure. By using this approach, our results are conservative and may understate the true returns to these strategies. This paper extends the research into the two investment approaches, by studying value and momentum jointly. Former research has indicated, that combining these two investment phenomena into one single strategy, may improve the returns. Using a simple market neutral strategy by investing 50 per cent in value and momentum respectively, our findings imply that this indeed has been the case on the Scandinavian stock market. The convincing Sharpe ratios for our combination strategy consistently beat both the value and momentum premiums. This is primarily due to an enhanced diversification effect attributable to a significant, negative correlation between value and momentum. Part of this advantageous synergy effect can be explained by the opposite signed exposure of value and momentum to size-factor. However, the 50/50 equal weighted combination of value and momentum is essentially immune to this size premium, which, contrary to traditional empirical findings and theory, has been negative on the Scandinavian stock market since 1991. Furthermore, our findings and derived conclusions are robust to a number of sensitivity analyzes, including but not limited to liquidity risk, financial stocks, the underlying national markets and recessions. Even though it may vary which of value and momentum are performing best, a combination of these seems to assure a stable and balanced investment for an investor. Additionally, we briefly indicate, that portfolio turnover and transaction cost should not affect our overall results and conclusion. However it may be necessary to omit a larger group of small and illiquid stocks. Finally, a more complex combination of value and momentum, using simultaneous ranking and classification of stocks is studied. Here, it is suggested, that buying cheap winners and selling expensive losers could improve the results even further. However, further investigation into these topics is beyond the scope of this paper.

EducationsMSc in Finance and Accounting, (Graduate Programme) Final Thesis
Publication date2014
Number of pages149