Comments on "the academic roots and evolution of real estate appraisal".
The authors' statement that "size and age alone will often explain 85% of the variation in selling price" after allocating for GIS (geographic information system) observations, got me wondering. Having developed an AVM for my home county, Bibb County, Georgia, I calculated the adjusted [R.sup.2] (the coefficient of determination, which indicates how much of the total value is explained by the model, with size and age being the only two independent variables. Since I basically duplicated the Realtors'[R] listing map in building the overall county model, I already had the county divided into 13 geographical economic markets; therefore, GIS factors were already considered to a certain extent. The resulting 13 coefficients of determination ranged from 15.9% to 82.8%, with their overall weighted average being 55%. Not bad considering the tremendously diverse markets that exist in our county. It was interesting to note that the most active one, North Macon-Bibb, with over 35% of all home sales, developed a coefficient of 76%. In other words, size and age alone explained over 75% of any home's market value within this submarket!
Because of these astounding (unbiased) results, I agree wholeheartedly with the authors' statement "... AVMs are likely to have a major impact on the demand for traditional appraisals over the next several years." However, in my opinion national AVMs (which cannot allocate for geographical submarkets as well as they need to) will soon be replaced by local AVMs developed by independent fee appraisers, who can develop their own AVM with a computer and Microsoft Office-Excel and who are accessible to any lender in the United States via the Internet.
Robert H. Gerhardt, MAI, SREA
Response--Comments on "The Academic Roots and Evolution of Real Estate Appraisal"
We greatly appreciate the interest and comments provided by Mr. Gerhardt. He is absolutely correct and we concur with all of his comments. We unintentionally used the word "independent" when we meant "dependent" with respect to estimating sales price and we appreciate the correction. He also makes a good point that locally developed AVMs will, like all appraisal work performed by those familiar with the local market, most likely be better because they will include nuances and important influences on value that national vendors may not discover.
Norman Miller, PhD, and Sergey Markosyan
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||letters to the editor|
|Author:||Gerhardt, Robert H.|
|Date:||Jul 1, 2003|
|Previous Article:||The Mystery of Capital: Why Capitalism Triumphs in the West and Fails Everywhere Else.|
|Next Article:||"A statistical definition of value": a critical comment.|