An Empirical Analysis of Hedonic Regression and Grid-Adjustment Techniques in Real Estate Appraisal

• Published in: Real estate economics Vol. 19; no. 1; p. 70
• Main Authors: Kang, Han-Bin, Reichert, Alan K
• Format: Journal Article
• Language: English
• Published: Bloomington Blackwell Publishing Ltd 01.03.1991
• Subjects: Economic models; Forecasting techniques; Multiple regression; Real estate appraisal; Regression analysis; State court decisions; Studies; United States > US
• ISSN: 1080-8620; 1540-6229
• DOI: 10.1111/1540-6229.00541

Multiple regression analysis (MRA) has become increasingly popular in appraising residential properties for tax purposes. At the same time, most fee appraisers and real estate brokers use the traditional sales comparison approach, sometimes referred to as the grid adjustment method. In a study, the 2 techniques are combined, using MRA to generate the adjustment coefficients used in the grid adjustment method. The combined grid-regression method is compared with ordinary regression. The results indicate that, in markets where equilibrium and housing and neighborhood characteristics are homogeneous, grid regression is the preferred technique. More specifically, the multiplicative percentage adjustment method is likely to be more accurate. In less homogeneous markets with significant price variation, the straight regression forecasting method using the log-linear functional form appears to be more appropriate. The use of ridge regression seems to be more appropriate in the presence of multicollinearity.