Ticks, Tweets, and Trails of Pain: Some Examples of Big Data in Business Research

• Activity Number: 201
• Type: Roundtables
• Date/Time: Monday, August 5, 2013 : 12:30 PM to 1:50 PM
• Sponsor: Section on Bayesian Statistical Science
• Author(s): James G Scott*+
• Companies: The University of Texas at Austin
• Keywords: Big data ; consumer sentiment ; financial risk ; causal inference

Abstract

In this talk, I will describe, at a fairly broad level, a few of the ways business researchers are using modern Big Data tools. I will focus on three specific case studies: (1) Ticks: how can we use tick-by-tick trading data to characterize the risk of a financial portfolio more accurately? (2) Tweets: how can we characterize consumer sentiment using data from social networks? (3) Trails of pain: how can we use the detailed case histories of hospital patients (e.g., insurance billing codes) in comparative-effectiveness research? I also will call attention to several issues, both statistical and computational, that are common to all these questions.