Predicting Litigation Risk via Machine Learning

45 Pages Posted: 11 Dec 2020

See all articles by Gene Moo Lee

Gene Moo Lee

University of British Columbia (UBC) - Sauder School of Business

James P. Naughton

University of Virginia, Darden School of Business

Xin Zheng

University of British Columbia

Dexin Zhou

City University of New York, Baruch College - Zicklin School of Business - Department of Economics and Finance

Date Written: December 1, 2020

Abstract

We demonstrate the value of machine learning in accounting through a detailed examination of litigation risk, an important and frequently used estimate in the literature. We evaluate a comprehensive set of twelve machine learning techniques and benchmark their performance against the logistic regression models in Kim and Skinner (2012). These models improve the prediction of litigation risk, with hourglass-shaped and convolutional neural networks the most effective. The improvements are substantial, and are driven by increased precision, the most salient attribute of litigation estimates in the accounting literature. We also produce firm-year litigation risk estimates for use in future research from a convolutional neural network model that uses recursive feature elimination on a pool of 68 possible parameters. Overall, our results suggest that the joint consideration of economically-meaningful predictors and machine learning techniques maximize the effectiveness of accounting estimates.

Keywords: Machine Learning, Securities Class Action Lawsuits, Neural Network

JEL Classification: G15, G18, M41

Suggested Citation

Lee, Gene Moo and Naughton, James P. and Zheng, Xin and Zhou, Dexin, Predicting Litigation Risk via Machine Learning (December 1, 2020). Available at SSRN: https://ssrn.com/abstract=3740954 or http://dx.doi.org/10.2139/ssrn.3740954

Gene Moo Lee

University of British Columbia (UBC) - Sauder School of Business ( email )

2053 Main Mall
Vancouver, BC V6T 1Z2
Canada

James P. Naughton

University of Virginia, Darden School of Business ( email )

P.O. Box 6550
Charlottesville, VA 22906-6550
United States

Xin Zheng (Contact Author)

University of British Columbia ( email )

2053 Main Mall
Vancouver, B.C. V6T 1Z2
Canada

Dexin Zhou

City University of New York, Baruch College - Zicklin School of Business - Department of Economics and Finance ( email )

55 Lexington Avenue
New York, NY 10010
United States

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
415
Abstract Views
1,454
rank
88,631
PlumX Metrics