Balistrieri, Thomas J.
Pavlov, Oleg V
The Central Bank of Russia has revoked 20% of all bank licenses since 2012. This project sought to determine a computational method for forecasting the probability of license revocation. To do this, we compared two classification algorithms, logistic regression and random forest. Using a Random Forest classifier produced higher F1 scores than logistic regression. We recommend that Deloitte continues the development of a random forest model by investigating clustering and additional features to improve the model’s performance.
Worcester Polytechnic Institute
Interactive Qualifying Project
All authors have granted to WPI a nonexclusive royalty-free license to distribute copies of the work, subject to other agreements. Copyright is held by the author or authors, with all rights reserved, unless otherwise noted.
Social Science and Policy Studies