CECL Validation Reports

A Current Expected Credit Loss (CECL) Data and Validation Report is accessible on the CECL Summary page for months with a Status of Complete or Passed.

Select the Download link in the Report column for a specific month on the CECL Summary page to download the report.

The report provides a summary of the quality and completeness of the data files transmitted to Deep Future Analytics (DFA).

Each active loan product is broken out and scored based on three data-quality components. The score for each component ranges from zero to five with five being the best score.

The data-quality components are:

Length

This score refers to the amount of historical data that is transmitted to DFA for the CECL calculation. The score considers the number of months of imported data that includes a loan pay-off and/or charge-off for each loan category. This rating builds up over time and requires at least 60 months of accumulated Passed data uploads to get a rating of five. A proportional rating is applied to any amount below 60.

Volume

This score is based on the average number of charge-offs per month imported into the model from the historical data files. The minimum number of charge-offs necessary to get a five-point rating for Volume is 10 per month. If the average number of charge-offs is below 10, then a proportional rating is applied.

Completeness

This score refers to a missing number of optional data fields within the files. The higher the score, the less data that is missing from the files.

Composite Score

This score is calculated from the individual component scores and is visually represented with a star rating. The Length and Volume components are weighted heavier than the Completeness component when determining the overall Composite Score.

The rest of the CECL Data and Validation Report provides recommendations on how to improve the quality ratings for the uploaded data along with a collection of tables and graphs summarizing the data that is provided for analysis.