Document Type

Thesis or dissertation

Date of this Version

2016

Abstract

After the Great Recession, there has been speculation as to whether it is possible to effectively detect systemic risk in the financial sector in order to guide macroprudential policy. This paper explores how aggregate indicators may be significant in forecasting banking crises among and across advanced economies. The paper begins by reviewing financial crisis theory and noteworthy qualitative frameworks and quantitative models for predicting financial crises. Mindful of the literature and models, machine learning techniques are used to assess the significance of 26 indicators in forecasting crises, two years in advance, for 20 high income countries. The classification models per country indicate that domestic credit to the private sector, as a percent of GDP, is the most common significant indicator in forecasting banking crises. When creating classification models inclusive of all countries, which are assumed to be of comparable financial depth, the bank lending-deposit spread becomes the most significant indicator. While the specificity of the models per country were quite high, the specificity dropped dramatically for models across countries. Overall, the results indicate a significant relationship between indicators of financial depth and banking crises, however, more data is needed to build upon these models to ensure their robustness.

Keywords

Forecasting, financial crises, systemic risk, macroprudential policy

Included in

Business Commons

Share

COinS
 

Date Posted: 09 August 2016

 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.