MdAPE is short for median absolute prediction error, and is a statistical measure of the valuation model accuracy. The valuation MdAPE is computed as follows:
Step 1: In any given month, HouseCanary produces and stores a set of 100m property valuations nationwide.
Step 2: Over the next month we obtain actual prices of arms length transactions occurring nationwide.
Step 3: We measure the absolute percent deviation of actual sales prices to what we had estimated the property values to be prior to the observed sale.
Step 4: Finally, the MdAPE is computed as the median absolute percent deviation of all transactions observed nationwide.
In summary, A lower MdAPE indicates a more accurate valuation.
The overall nationwide MdAPE is based on sales observed over the previous trailing 6 months. Recent monthly numbers provide insight into where the overall MdAPE is headed as we continue to collect more data and refine our property valuation model. Ideally, numbers for the most recent months should come in less than or equal to the overall MdAPE. This would imply that our overall MdAPE will continue to decrease as older, less accurate, estimates fall out of the 6 month trailing test window.
A lower MdAPE indicates a more accurate property valuation.
Residential real estate sales prices are influenced by much more than recent sales prices of nearby homes.
We pull sales and listing data from MLS and county assessor records, and pull insights from multiple other sources such as macroeconomic data, capital markets data, mortgage records, search and social data, and house/parcel data.
We run 8 individual AVMs, each of which executes different property valuation models. Every day, we use our unique ensemble methodology to run thousands of simulations for each property we track. We customize the weight we assign to various data sources to account for regional differences in market trends, economic factors, and demographics.
On a monthly basis, we compare our previous property valuations against the actual reported sales prices of homes. This allows us to continually evaluate the accuracy of our property valuations, identify changes in local and regional markets, and tune our models and algorithms to account for local and regional changes.