Wilary Winn CECL Modeling Proven Highly Predictive for Third Consecutive Year

We are proud to announce that for the third year in a row, our backtesting results showed that the Wilary Winn CECL model is highly predictive. Specifically, our results for the most recent three years are:

These results show that our model produces a modest conservative bias of 2 basis points and an average relative error of roughly 4%, with an overall accuracy of approximately 96%.

Moreover, we recently engaged Doeren Mayhew to validate our CECL model and framework to ensure they meet all regulatory requirements and accounting standards. The result was unequivocal: no findings, no required changes, and no weaknesses identified across model design, methodology, data handling, assumptions, or governance. Doeren Mayhew concluded that the model framework is conceptually sound, appropriately implemented, and fully fit for purpose under regulatory expectations.

Our ability to develop a highly predictive model within a robust framework makes us ideally suited to validate your CECL model and to provide an expert second opinion on the credit risk within your portfolio.

Let us know if you would like more information regarding our ongoing CECL modeling or CECL validation services or a copy of the Doeren Mayhew report.