On the Weights for Characteristics and Comparables for Property Valuation using Quality Rating Valuation Estimation

Abstract

This study considers the problem of finding weights for building characteristics and compares buildings in property valuation to provide a more rigorous analytical foundation for a simple yet practical valuation technique knowns as Quality Rating Valuation Estimation (QRVE). Mathematically, we prove that the “best” characteristic weights can be obtained from Multiple Linear Regression Analysis (MRA) coefficients. Furthermore, by applying the Gower Similarity index and the Partition Around Medoid (PAM) clustering technique, the proposed algorithm provides an appropriate similarity of the weighing of compared buildings. The case studies illustrate a way to select a subset of characteristics when there are many of them with two numerical examples, as well as a complete modification of QRVE in conjunction with the grid adjustment technique. The modified QRVE proposal results in a very reasonable and high valuation performance of the building value estimate.