Can Machines Learn Capital Structure Dynamics?

59 Pages Posted: 22 Oct 2019 Last revised: 30 Oct 2020

See all articles by Shahram Amini

Shahram Amini

University of Denver

Ryan Elmore

University of Denver - Daniels College of Business

Özde Öztekin

Florida International University (FIU)

Jack Strauss

University of Denver - Daniels College of Business

Date Written: October 29, 2020

Abstract

Yes, they can! Machine learning models that exploit big data identify leverage determinants and predict leverage better than classical methods. By allowing for nonlinearities and complex interactions, machine learning boosts the out-of-sample R-squared from 36% to 56% over linear methods such as LASSO. The best performing model (random forests) selects market-to-book, industry median leverage, cash & equivalents, Z-Score, profitability, stock returns, and firm size as reliable predictors of market leverage. Improved target measurement through machine learning yields 10%-34% faster adjustment relative to LASSO. Machine learning identifies uncertainty, cash flow, and macroeconomic considerations among primary drivers of leverage adjustments.

Keywords: Machine Learning, Target Leverage, Speed of Leverage Adjustment

JEL Classification: G0, G17, G30, G32, C10, C50

Suggested Citation

Amini, Shahram and Elmore, Ryan and Öztekin, Özde and Strauss, Jack, Can Machines Learn Capital Structure Dynamics? (October 29, 2020). Available at SSRN: https://ssrn.com/abstract=3473322 or http://dx.doi.org/10.2139/ssrn.3473322

Shahram Amini (Contact Author)

University of Denver ( email )

2101 S. University Blvd
Denver, CO 80208
United States

HOME PAGE: http://https://daniels.du.edu/directory/shahram-amini/

Ryan Elmore

University of Denver - Daniels College of Business ( email )

2101 S. University Blvd.
Denver, CO 80208
United States

Özde Öztekin

Florida International University (FIU) ( email )

University Park
11200 SW 8th Street
Miami, FL 33199
United States

Jack Strauss

University of Denver - Daniels College of Business ( email )

2101 S. University Blvd.
Denver, CO 80208
United States

Here is the Coronavirus
related research on SSRN

Paper statistics

Downloads
1,164
Abstract Views
2,111
rank
19,923
PlumX Metrics