Investigating bank failures using text mining

Aparna Gupta, Majeed Simaan, Mohammed J. Zaki

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

We extend beyond healthiness assessment of banks using quantitative financial data by applying textual sentiment analysis. Looking at public annual reports for a large sample of U.S. banks in the 2000-2014 period, we identify 52 public bank holding companies that were associated with bank failures during the global financial crisis. Utilizing sentiment dictionaries designed for financial context, we find that negative and positive sentiments discriminate between failed and non-failed banks 88% and 79%, respectively, of the time. However, we find that positive sentiment contains stronger predictive power than negative sentiment; out of ten failed banks, on average positive sentiment can identify six true events, while negative sentiment identifies five failed banks at most. While one would link financial soundness with more positive sentiment, it appears that failed banks exhausted more positive sentiment than their non-failed peers, whether ex-ante in anticipation of good news or ex-post to conceal financial distress.

Original languageEnglish
Title of host publication2016 IEEE Symposium Series on Computational Intelligence, SSCI 2016
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781509042401
DOIs
Publication statusPublished - 9 Feb 2017
Event2016 IEEE Symposium Series on Computational Intelligence, SSCI 2016 - Athens, Greece
Duration: 6 Dec 20169 Dec 2016

Other

Other2016 IEEE Symposium Series on Computational Intelligence, SSCI 2016
CountryGreece
CityAthens
Period6/12/169/12/16

Fingerprint

Text Mining
Glossaries
Industry
Sentiment Analysis
Financial Crisis
Anticipation
Financial Data
Banks
Bank failure
Text mining
Sentiment
Soundness
Annual

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Information Systems and Management
  • Control and Optimization
  • Artificial Intelligence

Cite this

Gupta, A., Simaan, M., & Zaki, M. J. (2017). Investigating bank failures using text mining. In 2016 IEEE Symposium Series on Computational Intelligence, SSCI 2016 [7850006] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/SSCI.2016.7850006

Investigating bank failures using text mining. / Gupta, Aparna; Simaan, Majeed; Zaki, Mohammed J.

2016 IEEE Symposium Series on Computational Intelligence, SSCI 2016. Institute of Electrical and Electronics Engineers Inc., 2017. 7850006.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Gupta, A, Simaan, M & Zaki, MJ 2017, Investigating bank failures using text mining. in 2016 IEEE Symposium Series on Computational Intelligence, SSCI 2016., 7850006, Institute of Electrical and Electronics Engineers Inc., 2016 IEEE Symposium Series on Computational Intelligence, SSCI 2016, Athens, Greece, 6/12/16. https://doi.org/10.1109/SSCI.2016.7850006
Gupta A, Simaan M, Zaki MJ. Investigating bank failures using text mining. In 2016 IEEE Symposium Series on Computational Intelligence, SSCI 2016. Institute of Electrical and Electronics Engineers Inc. 2017. 7850006 https://doi.org/10.1109/SSCI.2016.7850006
Gupta, Aparna ; Simaan, Majeed ; Zaki, Mohammed J. / Investigating bank failures using text mining. 2016 IEEE Symposium Series on Computational Intelligence, SSCI 2016. Institute of Electrical and Electronics Engineers Inc., 2017.
@inproceedings{98d0c8ac42624b27bd831d136f33774f,
title = "Investigating bank failures using text mining",
abstract = "We extend beyond healthiness assessment of banks using quantitative financial data by applying textual sentiment analysis. Looking at public annual reports for a large sample of U.S. banks in the 2000-2014 period, we identify 52 public bank holding companies that were associated with bank failures during the global financial crisis. Utilizing sentiment dictionaries designed for financial context, we find that negative and positive sentiments discriminate between failed and non-failed banks 88{\%} and 79{\%}, respectively, of the time. However, we find that positive sentiment contains stronger predictive power than negative sentiment; out of ten failed banks, on average positive sentiment can identify six true events, while negative sentiment identifies five failed banks at most. While one would link financial soundness with more positive sentiment, it appears that failed banks exhausted more positive sentiment than their non-failed peers, whether ex-ante in anticipation of good news or ex-post to conceal financial distress.",
author = "Aparna Gupta and Majeed Simaan and Zaki, {Mohammed J.}",
year = "2017",
month = "2",
day = "9",
doi = "10.1109/SSCI.2016.7850006",
language = "English",
booktitle = "2016 IEEE Symposium Series on Computational Intelligence, SSCI 2016",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Investigating bank failures using text mining

AU - Gupta, Aparna

AU - Simaan, Majeed

AU - Zaki, Mohammed J.

PY - 2017/2/9

Y1 - 2017/2/9

N2 - We extend beyond healthiness assessment of banks using quantitative financial data by applying textual sentiment analysis. Looking at public annual reports for a large sample of U.S. banks in the 2000-2014 period, we identify 52 public bank holding companies that were associated with bank failures during the global financial crisis. Utilizing sentiment dictionaries designed for financial context, we find that negative and positive sentiments discriminate between failed and non-failed banks 88% and 79%, respectively, of the time. However, we find that positive sentiment contains stronger predictive power than negative sentiment; out of ten failed banks, on average positive sentiment can identify six true events, while negative sentiment identifies five failed banks at most. While one would link financial soundness with more positive sentiment, it appears that failed banks exhausted more positive sentiment than their non-failed peers, whether ex-ante in anticipation of good news or ex-post to conceal financial distress.

AB - We extend beyond healthiness assessment of banks using quantitative financial data by applying textual sentiment analysis. Looking at public annual reports for a large sample of U.S. banks in the 2000-2014 period, we identify 52 public bank holding companies that were associated with bank failures during the global financial crisis. Utilizing sentiment dictionaries designed for financial context, we find that negative and positive sentiments discriminate between failed and non-failed banks 88% and 79%, respectively, of the time. However, we find that positive sentiment contains stronger predictive power than negative sentiment; out of ten failed banks, on average positive sentiment can identify six true events, while negative sentiment identifies five failed banks at most. While one would link financial soundness with more positive sentiment, it appears that failed banks exhausted more positive sentiment than their non-failed peers, whether ex-ante in anticipation of good news or ex-post to conceal financial distress.

UR - http://www.scopus.com/inward/record.url?scp=85016053036&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85016053036&partnerID=8YFLogxK

U2 - 10.1109/SSCI.2016.7850006

DO - 10.1109/SSCI.2016.7850006

M3 - Conference contribution

BT - 2016 IEEE Symposium Series on Computational Intelligence, SSCI 2016

PB - Institute of Electrical and Electronics Engineers Inc.

ER -