The extent to which an AVM’s valuations approximate actual market values. Analysts usually examine a model’s overall accuracy through two different types of metrics – measures of AVM bias and measures of AVM precision.
The "Confidence Scores" provided with most AVMs are not standardized, but they are a very effective guideline estimating the accuracy and precision of AVM value estimates.
AVMs are great, but using them properly requires being careful to follow the rules set by the OCC. This paper outlines how to outsource AVM validation and stay in "compliance" with regulators' requirements.
We have mixed feelings about Mark Sennott's whitepaper. He echoes our concerns about why AVM testing is important, but he reveals that AVM test results are NOT determining which AVMs are used.