Effective January 1, 2020, Fannie Mae introduced a new standard for lenders who use statistical sampling to select loans for QC. Highlighted details can be found in the updated Selling Guide, on page 1070:
“If the lender uses statistical sampling for its selection process, at a minimum, the statistical sampling model (variables) must be calculated using a 95% confidence level with a 2% precision rate and a statistical statement of six months (Fannie Mae recommends using a three month statement.)”
What does this mean for you? How do you calculate the correct number of loans to sample and audit under the new standard? Cogent’s statistical sample size calculator makes it easy:
To calculate a six month statement – Enter the TOTAL estimated population for the 6 months (usually by extrapolating current origination levels). Then DIVIDE the calculated Sample Size by 6 to derive a sample size for a single month.
To calculate a three month statement – Enter the TOTAL estimated population for the 3 months (usually by extrapolating current origination levels). Then DIVIDE the calculated Sample Size by 3 to derive a sample size for a single month.
Thus, a lender originating an average of 1,000 loans per month might estimate a 6-month population of 6,000 loans. Assuming an expected incidence rate (or defect rate) of 5% and a precision target of 2%, the resulting Sample Size becomes 242. Dividing this by 6 months yields a monthly sample size of 40 loans.
For more information, please see the following:
– The article “Sample Examples – The Calculator in Action” provides guidance for using the Calculator below in various scenarios.
– For an explanation of Cogent’s sampling methodology, please see the white papers on our Resources page.