An automated valuation model (AVM) provides an estimate of market value for a subject property at a specific point in time. To determine a value for the subject property, the most common AVM method is the use of prices of comparable properties that were recently sold. This is typically done by identifying all properties within a certain distance of the subject and then applying guidelines to find those that are most similar to the subject in terms of property characteristics and location factors.
This is a similar methodology of social networks, such as Facebook and LinkedIn: people are connected based on similarity of relationships. As an example, a person might be connected with his or her colleagues or someone they met at a conference, with the strength of a linkage dependent upon how many connections they have in common.
Greater Similarity…Stronger Connection
Applying this concept to the housing market, it is possible to construct a network of properties—a network graph algorithm. Nodes on the graph represent individual properties, and the strength of each linkage indicates the similarity of connected properties. A linkage can also be built on factors beyond property characteristics. Consider the example of a property used as a comparable sale by an appraiser for another property; it indicates that the two are deemed similar to each other, therefore the linkage between the pair would be very strong.
CoreLogic data sources can be leveraged to build a network of all properties in the nation, making the selection of comparable properties straightforward, with the best ones being those with the strongest affiliation to the subject. The accompanying illustration of a network graph shows a connection between a subject (red node) and its “comparables” (blue nodes); all the connections are linked either directly or indirectly.
For a better conceptualization of a network graph, consider placing these interconnected data points on a selected area in Google Maps. In this example, the red pin represents the subject property, pink circles indicate comparable properties that were previously used by an appraiser, and blue pins are ideal comparables identified by the CoreLogic network graph algorithm.
Upon analysis of the map, the first takeaway is that the CoreLogic algorithm picks up all but one of the appraiser’s comparables. This is an indicator of success of the comparable selection, as nobody knows a neighborhood better than a local appraiser. The second conclusion is that comparables tend to be in the same neighborhood as the subject, or in the same general vicinity as those previously selected by the appraiser.
Single Model Methodology Is the Single Best Approach
Leveraging a network graph algorithm to select comparables the same way a trained local expert does is just one of the innovations driving CoreLogic’s new Total Home ValueX AVM.
This recently introduced valuation solution leverages artificial intelligence and machine learning capabilities built on Cloud technology. Using single model methodology to support diverse use cases and markets, users only need to validate one model, which is tuned by use case to deliver unsurpassed hit rates, accuracy and consistency.
With access to a property database of more than 5.5 billion records (updated daily) that captures 99.9% of U.S. properties and spans more than 50 years, THVx produces automated valuations that can be used anywhere a current property value is relevant.