Momentum Investing and Stock Return Predictability: An Empirical Study on S&P 500

Abstract

This thesis aims to examine possible ways of predicting stock returns. In addition, this thesis examines whether a momentum investing strategy would have yielded a positive return between 2000 and 2021 using the S&P 500 index. Furthermore, it is sought to investigate which factors influence market value and stock return, hence assessing the possibility of predicting stock returns and company value using different financial ratios and metrics. Previous research has proved the existence of the momentum effect. In contrast, this thesis seeks to cover a gap in the research by investigating more recent data from the period between 2000 and 2021 and hence whether the momentum effect has continued to exist, using data from the EikonDatastream database to analyze both the momentum effect and to predict stock return using Excel and R to process the data. The findings of this thesis prove that by investing in past winning stocks, it is possible to obtain an excess return above the S&P index using both a long strategy and an intermediate strategy. The findings can partly be explained by combining the Capital Asset Pricing Model and behavioral finance theory. Through a multiple regression analysis using OLS, it is possible to create a model which is better at predicting market values than a model guessing on the mean, although some assumptions of OLS are violated. Thus, the implications of this thesis are a more present-day perspective of the impact of momentum investment strategies and stock return predictability