CoCo Seminar: "Creating a Voter Sentiment Index by Reading Tweets" by Wei Xiao
Details
Dr. Wei Xiao
Professor, Department of Economics
Binghamton University
Wednesday February 24, 2021
11:00am-12:00pm
Online (via Zoom)
https://binghamton.zoom.us/j/97329090945?pwd=WEdJb3ZWczhYcDlYZDRzOGdvdy9kUT09
Abstract:
We use natural language processing to read hundreds of thousands of tweets every day, and analyze how voters' sentiment towards Biden and Trump changed between September 4 and November 3, 2020. Preliminary results show that the indices clearly responded to news, and correlated with changes in average national polls. In most of the survey period, the sentiment was more pro-Biden. However, the differences vanished approaching election day. Verified influencers were more pro-Biden, while unverified influencers were much more pro-Trump.
Dr. Wei Xiao is Professor and Director of Undergraduate Studies in the Department of Economics at Binghamton University. He received his Ph.D. in Economics at the University of Pittsburgh. His research interests include macroeconomics, behavioral macroeconomics, adaptive learning, monetary theory and policy, and macro aspects of international economics.
For more information, contact Hiroki Sayama (sayama@binghamton.edu) or Andreas Pape (apape@binghamton.edu).
http://coco.binghamton.edu/