Document Type
Thesis
First Faculty Advisor
A. Can Inci
Second Faculty Advisor
Kevin Mentzer
Keywords
asset pricing; sentiment; finance; investments
Publisher
Bryant University
Rights Management
CC 4.0 BY-NC-SA
Abstract
This paper aims to explore the influence of daily news article sentiment as a predictor of the returns on an investment. A daily sentiment score is developed and it is used to augment five commonly researched models, CAPM, Fama & French Three Factor, Carhart Four Factor, Fama & French Five Factor, and Fama & French Five Factor with Momentum. This study looks at six different securities over a five-year period. Along with this, two different variations of the factor are looked at, one of which is a simple factor that scores days without an article as zero and another that attempts to use a reduced value of previous scores instead of zero. Overall, the research finds little evidence between the developed factors and the return of an investment. In the end it is suggested to use multiple news sources, and a more precise sentiment scoring process in order to increase the accuracy of the sentiment factor and the study.
