Extending the Logit Model with Midas Aggregation: The Case of US Bank Failures

30 Pages Posted: 16 Feb 2018  

Francesco Audrino

University of St. Gallen

Alexander Kostrov

University of St. Gallen

Juan-Pablo Ortega

Universität Sankt Gallen; Centre National de la Recherche Scientifique (CNRS)

Date Written: February 4, 2018

Abstract

We propose a new approach based on a generalization of the classic logit model to improve prediction accuracy in US bank failures. We introduce mixed-data sampling (Midas) aggregation to construct financial predictors in a logistic regression. This allows us to relax the limitation of conventional annual aggregation in financial studies. Moreover, we suggest an algorithm to reweight observations in the log-likelihood function to mitigate the class-imbalance problem, that is, when one class of observations is severely undersampled. We also address the issue of the classification accuracy evaluation when imbalance of the classes is present. In applying the suggested model to the period from 2004 to 2016, we show that it correctly classifies more bank failure cases than the reference logit model introduced in the literature, in particular for long-term forecasting horizons. This improvement has a strong significant impact both in statistical and economic terms. Some of the largest recent bank failures in the US that were previously misclassified are now correctly predicted.

Keywords: bank failures, prediction, mixed-data sampling, logit model

JEL Classification: C38, C53, G21

Suggested Citation

Audrino, Francesco and Kostrov, Alexander and Ortega, Juan-Pablo, Extending the Logit Model with Midas Aggregation: The Case of US Bank Failures (February 4, 2018). Available at SSRN: https://ssrn.com/abstract=3117877 or http://dx.doi.org/10.2139/ssrn.3117877

Francesco Audrino

University of St. Gallen ( email )

Bodanstrasse 6
St. Gallen, CH-9000
Switzerland

Alexander Kostrov (Contact Author)

University of St. Gallen ( email )

Bodanstrasse 6
St. Gallen, CH-9000
Switzerland

Juan-Pablo Ortega

Universität Sankt Gallen ( email )

Bodanstrasse 6
St. Gallen, St. Gallen CH-9000
Switzerland

HOME PAGE: http://juan-pablo-ortega.com

Centre National de la Recherche Scientifique (CNRS) ( email )

16 route de Gray
Besançon, 25030
France

HOME PAGE: http://juan-pablo-ortega.com

Register to save articles to
your library

Register

Paper statistics

Downloads
59
rank
336,070
Abstract Views
232
PlumX