MIT EECS | Hudson River Trading Undergraduate Research and Innovation Scholar
Modeling and Predicting Investor Trading Behavior Using Media Sentiment
- Artificial Intelligence and Machine Learning
Andrew W. Lo
The rapid integration of information sharing platforms has dramatically impacted the financial markets and sparked many questions about the relationship between media sentiment and alpha signals. Current research investigates the effects of sentiment on liquidity, stock returns, and market movement. To the best of our knowledge, there is no study that explores the relationship between market sentiment and individual trading behavior. In this study, we will investigate the impact of news sentiment on investor trading behavior with a large novel data set of brokerage accounts and RavenPack News Analytics data. We will extract features from trading activity for each investor, compute aggregate sentiment scores for each time period, and investigate correlations between the two. Then, we will attempt to predict how individuals will react to changes in market sentiment.
” I am excited to gain insight into a wide variety of topics during guest lectures and experience the research process firsthand throughout this SuperUROP project. Moreover, I hope to increase my understanding of financial economics, develop my skills in data analytics, and apply machine learning concepts from class to the real world.”