Comparable Weighting and Selection in Real Estate Valuation: A Hedonic Regression Approach

Introduction – Comparable evidence is at the heart of virtually every real estate valuation and is fundamental in producing a sound estimate. Although valuers are experienced in handling and analyzing comparable data, approaches vary with many degrees of freedom potentially leading to incorrect decisions. And even with the increasing amount of computer-driven estimates, the evidence of comparable transactions remains paramount.Purpose – This paper formalizes certain aspects of selecting comparables. We apply economic theory to select criteria that make properties comparable, determine how comparables can be weighted and ultimately, make decisions about what comparables should be reported such that the estimate is both unbiased and of minimum variance. In addition to this quantitative optimum, we hold a survey among real estate experts to test qualitative how well the model outcomes translate back to practice to gain additional insights where and why misalignment might occur.Design/methodology/approach – We start with an extensive literature review about the work that has previously been done in this space. We then apply hedonic price theory to estimate the weights of the comparables and construct a novice comparability rater model. Finally, we briefly evaluate how these results compare to other promising methodologies found in the literature review and survey these among expert valuers.Originality/value – To our knowledge, this paper is the first to cover a critical overview of work that has been done in this field. Furthermore, the application of established hedonic price theory for the purpose of weighting comparables deserves a new light within the abundance of 'newer' complex approaches. Lastly, since the relevance of this research is more practical, the translation of the results back to the experts is not found in previous work.Findings – None yet (finish end of April)