Modeling Apartment Characteristic Valuations

• Activity Number: 423
• Type: Contributed
• Date/Time: Tuesday, August 3, 2010 : 2:00 PM to 3:50 PM
• Sponsor: Business and Economic Statistics Section
• Author(s): Caroline Swiger*+ and Julia Sharp and William Bridges
• Companies: Clemson University and Clemson University and Clemson University
• Keywords: hedonic regression ; regression

Abstract

The objective of this study was to consider a regression model to estimate the actual value of apartment characteristics in a metropolitan area (i.e., hedonic regression). Many stated values of characteristics are based on professional judgment, competitors' rents, and subjective opinions. Regression modeling is used to estimate actual values, using apartment rent as the dependent variable and apartment characteristics as independent variables. A comparison of stated and actual values is used to determine the best form of the regression model and model fit. Forms of the independent indicator variables considered are 1) each variable modeled separately, 2) all variables, and 3) all variables including interactions. Preliminary analyses revealed large discrepancies between actual and stated values of apartment characteristics for each of the three different models appraised.