A Picture is Worth a Thousand Words: Measuring Investor Sentiment by Combining Machine Learning and Photos from News

74 Pages Posted: 10 Dec 2018 Last revised: 2 Jul 2020

See all articles by Khaled Obaid

Khaled Obaid

California State University-East Bay

Kuntara Pukthuanthong

University of Missouri, Columbia

Date Written: June 29, 2020

Abstract

By applying machine learning to accurately and cost effectively classify photos based on sentiment, we introduce a daily market-level investor sentiment index (Photo Pessimism) from a large sample of news photos. Between 1926 and 2018, Photo Pessimism predicts market return reversal and increase in trading volume. The predictability pattern is concentrated among stocks with high limits to arbitrage and during high uncertainty periods. Photo Pessimism enhances pessimism embedded in text, but subsumes it during days with traumas and influential photos. Photo Pessimism from business news has over six times stronger predictive power than Photo Pessimism from general news.

Keywords: Investor Sentiment, Behavioral Finance, Return Predictability, Machine Learning, Big Data

JEL Classification: C53, G10, G17

Suggested Citation

Obaid, Khaled and Pukthuanthong, Kuntara, A Picture is Worth a Thousand Words: Measuring Investor Sentiment by Combining Machine Learning and Photos from News (June 29, 2020). Available at SSRN: https://ssrn.com/abstract=3297930 or http://dx.doi.org/10.2139/ssrn.3297930

Khaled Obaid

California State University-East Bay ( email )

435 Valley Business and Technology Center
College of Business and Economics
Hayward, CA 94621
United States

HOME PAGE: http://sites.google.com/alumni.wfu.edu/khaledobaid/

Kuntara Pukthuanthong (Contact Author)

University of Missouri, Columbia ( email )

Robert J. Trulaske, Sr. College of Business
403 Cornell Hall
Columbia, MO 65211
United States
6198076124 (Phone)

HOME PAGE: http://https://kuntara.weebly.com

Here is the Coronavirus
related research on SSRN

Paper statistics

Downloads
465
Abstract Views
3,711
rank
68,541
PlumX Metrics