Online Program

Friday, February 20
CS06 Business Applications Fri, Feb 20, 11:00 AM - 12:30 PM
Borgne

Applying Econometric Time Series Methods to CCAR Requirements (302960)

Christian Macaro, SAS Institute  
*Kenneth Sanford, SAS Institute 

Keywords: economics, econometrics, banking, risk, CCAR, stress tests

U.S. banks and bank-holding companies have faced increased regulation following the financial crisis of 2008. One of the recent changes to these regulations is known as CCAR, or Comprehensive Capital Analysis and Review. At the core of these new regulations are a series of “what-if ” or “scenario analysis” requirements involving a number of Federal Reserve–provided scenarios. We will describe the CCAR regulations and propose frequentist and Bayesian econometric time series methods to deal with this stress-testing problem from a highly practical “top-down” form. The topic will focus on the value of using univariate and multivariate econometric time series methods, as well as the methodology behind these models.