Comparative Analysis of Extreme Winner & Loser Stocks

Frederik Nordgaard Hansen & Christoffer Holmeskov Rasmussen

Student thesis: Master thesis

Abstract

This paper aims to challenge the efficient market hypothesis by evolving a strategy based on historical firm-specific characteristics. The paper has drawn inspiration from Marc Reinganum’s (1988) article “The Anatomy of Stock Market Winner”. The article characterizes extreme stock winners on the American stock market in the period 1970-1983 and implements two strategies that outperform the market. As opposed to Reinganum’s article, this paper will include extreme losers to distinguish winners from losers in order to improve the strategies. This paper examines 349 extreme stock winners and 163 extreme stock losers in the period of 2000- 2017 in the American stock market. The extreme stocks are tested based on 35 different parameters with four different testing methods. In order to distinguish winners from losers, each method of testing is examined with a statistic t-test and z-test. To prevent the phenomenon of data mining each of the characteristics is justified with economic and/or empirical evidence. The eight most significant parameters have been used to formulate four different investment strategies. The four strategies are tested against the Nasdaq Composite Index, and all four strategies outperform the market out-of-sample based on Sharpe Ratio. The performance could not entirely be explained by CAPM or the three-factor model, why this paper challenges the efficient market hypothesis. Finally, this paper discusses potential biases and explanations as to how it is possible to outperform the market. The potential biases from the quantitative analysis include data mining bias, look-ahead bias, survivorship bias, and subjectivity in the methods and how to avoid it. Furthermore, traditional financial theory and behavioral finance are leveraged to gain their perspective on the performance. Traditional financial theory argues that risk has not been taken into account, such as bankruptcy and liquidity risk. Behavioral finance argues that investors act irrationally causing opportunities to outperform the market. Lastly, transaction cost and its influence on performance is discussed. This paper concludes that the four strategies outperform the market in spite of being subject to biases and risks that are not taken into account. Conclusively, it is recommended to make a further analysis of the potential biases, risks, and costs before implementing the strategies.

EducationsMSc in Finance and Accounting, (Graduate Programme) Final Thesis
LanguageDanish
Publication date2021
Number of pages124
SupervisorsThomas Plenborg