Enhancing value strategy with momentum on US stock markets

Abstract

Markets are often assumed to be efficient. According to efficient market hypothesis all relevant information is instantly reflected to the prices of securities and funds are allocated to the most profitable investments. Changes in prices happen randomly, and there should not exist any regularities on the markets. However, regularities still exist and for many of these, none generally accepted explanation have been found.

The aim of this master’s thesis is to investigate two different anomalous investment strategies; value strategy and momentum strategy. Value strategy was found during the first half of 20th century, and excess returns have been achieved with it for decades. Momentum anomaly was found in the 1990’s and likewise, excess returns have been achieved with it. In this master’s thesis, the performance of value strategy on US stock markets is analyzed during the 21st century and additionally, could the return of value strategy be enhanced with momentum.

Not all value stocks are automatically profitable. Behind the theory is that the price of a value stock is in the bottom of the stock’s price cycle. A value stock that has entered the uprising part of its price cycle could be recognized with momentum. Portfolios used in this thesis are formed based on valuation ratios of price-to-earnings, price-to-book and a combination of enterprise-value-to-earnings-before-interest-taxes-depreciatio-and-amortization and price-to-sales. Portfolios are held with different holding periods, because according to earlier studies, the optimal holding period for value strategy and momentum strategy are not similar. Total amount of portfolios analyzed in this thesis is 108. Jensen’s alpha is calculated by using Carhart’s four-factor-model, which includes risk factors for value and momentum.

Portfolios which contained value stocks performed better compared to other portfolios with all valuation methods and holding periods. Including momentum to portfolio formation did not enhance the returns noticeably, except with the use of combination of EV/EBITDA and P/S as valuation method and with one year holding period. Only three portfolios produced statistically significant alfa which were all formed by the above-mentioned formation criteria. Carhart’s four-factor-model explained the returns well. Possible topics for further research is portfolio formation by different kind of combinations of value ratios and including momentum indicator into those portfolios.