Kan man slå markedet?

Julie Helm Trige Starklit & Emil Vestergaard Jensen

Student thesis: Master thesis


The objective of this thesis is to examine whether it is possible to obtain an abnormal return by investing in stocks that are temporarily beaten down. The thesis is divided into five parts. The first part is the motivation for the project as well as a presentation of the problem statement. The second part is an examination of the methods that are being applied to answer the problem statement. Third, the underlying theory is presented to get a theoretical foundation behind the two different views on the financial market as well as the measures that are being used in the tests. The fourth part contains an analysis of the test results. Lastly, the results have been interpreted into a discussion to explain the results more thoroughly. The main findings of the thesis have been summarised in a conclusion, and suggestions for further research are presented at the end. To answer the problem statement the project includes a test that examines whether certain key measures can be used to identify beaten down stocks that are undervalued but will return to their fundamental value within a few years. The CAPM model has been used to calculate residual returns of the sample, which is composed of the Russell 3000 index. The residual return is applied as a measure of the stock returns. The method used in the test is linear regression. Each key measure is tested against its impact on the stock returns of the sample. The significant key measures are then tested in multiple regressions with the purpose of finding the most significant regression. The coefficients of this regression will be used as a basis to rate stocks in another sample. If the screening tool works, the stocks with the highest ratings will be the ones with the highest returns and the stocks with the lowest ratings will have the lowest returns. For a better understanding, it has been necessary to examine what different theories say about whether or not it is even possible to outperform the market. For this reason, Fama’s Efficient Market Hypothesis has been included, as he argues that the markets are efficient and thus it is not possible to beat the market systematically. Behavioural Finance is included as counterargument to Fama, as theories within this area suggest that pricing biases in the stock market indicate that markets are not efficient. In short, Fama argues that when arbitrage opportunities occur as a result of mispricing, rational investors will exploit these and thus push the prices of companies towards their fundamental values. However, the Behavioural Finance theory states that both irrational investors, as well as Limits to Arbitrage, can result in pricing biases that can both be significant and persistent. Based on the debate between the two schools, it has been assessed that it is worthwhile to Executive Summary Side 2 af 114 test if it is possible to identify mispricing of companies by using company specific key measures. To create a screening tool, it has been necessary to find theoretical and empirical evidence about what key measures should be used in the test, which will also take place in this part of the thesis. In the fourth part of the thesis, the test and analysis have been carried out. Through simple linear regression, it was found that the ratios P/E, P/B, FCF yield and Leverage all have significant impact on the stock returns despite poor explanatory power. All the significant ratios have been tested in multiple regressions. The regression that has turned out to be the most significant comprises FCF yield, Leverage and P/B. As part of the analysis, it has been tested if the results are externally valid. This has been tested on two out-of-samples: one in the same time period but containing different companies and one in a different period of time. The out-of-sample results are not significant, which implies that the model is not externally valid. The reasoning for this is discussed more thoroughly in the fifth section of the thesis. The discussion has focused on the fact that it seems that macro economic factors have an impact on the results. This is due to the fact that high multiples show a negative impact on the stock return in one time period while they show a positive impact in another. The macro economic impact is closely related to the investor behaviour, since investors seem to be short sighted and thus changes their risk preferences when the economy changes. This has indicated that it could be necessary to include macro factors in the model to end up with a significant screening tool. However, the results can also be influenced by randomness and thus indicate that the market is efficient. The discussion also dealt with the fact that transaction and information costs are not as high as they used to be. This is an indication of the markets possibly being more efficient than earlier. No matter if it is due to the model being inadequate or the market being efficient, the conclusion must be that the desired screening tool has not been possible to create. The conclusion of the thesis is that according to the method it is not possible to systematically outperform the market, and thus it is not possible to obtain an abnormal return by investing in stocks that are temporarily beaten down. Suggestions for further research are presented in the end of the thesis. Suggestions have been made about working further with the problem by including macro economic factors and testing if the market is efficient by using different measures.

EducationsMSc in Finance and Accounting, (Graduate Programme) Final Thesis
Publication date2015
Number of pages140