Date: Wednesday February 26, 2013
Time: 6:00 pm
30 Arbor St, Hartford, CT
Can the choice of words and tone used by the authors of financial news articles correlate to measurable stock price movements? If so, can the magnitude of price movement be predicted using these same variables? We investigate these questions using the Arizona Financial Text (AZFinText) system, a financial news article prediction system, and pair it with a sentiment analysis tool. Through our analysis, we found that subjective news articles were easier to predict in price direction (59.0% versus 50.0% of chance alone) and using a simple trading engine, subjective articles garnered a 3.30% return. Looking further into the role of author tone in financial news articles, we found that articles with a negative sentiment were easiest to predict in price direction (50.9% versus 50.0% of chance alone) and a 3.04% trading return. Investigating negative sentiment further, we found that our system was able to predict price decreases in articles of a positive sentiment 53.5% of the time, and price increases in articles of a negative sentiment 52.4% of the time. We believe that perhaps this result can be attributable to market traders behaving in a contrarian manner, e.g., see good news, sell; see bad news, buy.
About the Speaker
Dr. Schumaker is an Associate Professor of Management Information Systems at Central Connecticut State University. He received his PhD in Management from the University of Arizona in 2007, an MBA degree in Management and International Business from the University of Akron in 2001, and a Bachelors of Science degree in Civil Engineering from the University of Cincinnati in 1997.
He is also the Managing Editor for the CIIMA journal, has authored a book on Sports Data Mining, several book chapters, multiple journal articles in DSS, ACM TOIS, CACM and JASIST as well as had his research featured in the Wall Street Journal and numerous other media outlets.
Dr. Schumaker’s overall research interests involve the uses of technology to acquire, deliver and make predictions in a variety of Business-related environments. These interests further branch into computer mediated communications, design science, human computer interfaces, machine learning algorithms, natural language processing, technology acceptance models and textual data mining. In particular, he focuses on the areas of Question/Answer systems, Textual/Financial prediction and Sports Data Mining.
About the Venue
This talk will be held at MakeHartford, 30 Arbor St, Hartford CT. Enter the building via the main door and take the elevator down 1 floor to the basement. U-turn off the elevator and take a left. We are in the first right door. If you get lost, follow the bright green signs or call 203-516-0077 and someone will assist you.