Prediction accuracy in mass appraisal: A comparison of modern approaches

William J. Mccluskey, M. McCord, P. T. Davis, M. Haran, D. McIlhatton

Research output: Contribution to journalArticle

31 Citations (Scopus)

Abstract

The advancement of computational software within the last decade has facilitated enhanced uptake of mass appraisal methodologies by the valuation and prediction accuracy in computer-assisted mass appraisal community for price modelling, estimation and tribunal defence. Applying a sample of 2694 residential properties, this paper assesses and analyses a number of geostatistical approaches relative to an artificial neural network (ANN) model and the traditional linear hedonic pricing model for mass appraisal valuation accuracy and price estimation purposes. The findings demonstrate that the geostatistical localised regression approach is superior in terms of model explanation, reliability and accuracy. ANNs can be shown to perform very well in terms of predictive power, and therefore valuation accuracy, outperforming the traditional multiple regression analysis (MRA) and approaching the performance of spatially weighted regression approaches. However, ANNs retain a 'black box' architecture that limits their usefulness to practitioners in the field. In relation to cost-effectiveness and user-friendly applicability for the valuation community, the MRA approach outperforms the 'black box' nature of the ANN technique, with the geographically weighted regression approach providing the best balance of outright performance and transparency of methodology. It is this spatially weighted approach utilising absolute location which appears to represent the way forward in developing the practice of mass appraisal.

Original languageEnglish
Pages (from-to)239-265
Number of pages27
JournalJournal of Property Research
Volume30
Issue number4
DOIs
Publication statusPublished - Dec 2013
Externally publishedYes

Fingerprint

valuation
prediction
regression
neural network
regression analysis
artificial neural network
multiple regression
methodology
transparency
community
performance
pricing
comparison
appraisal
costs
software
cost
modeling
price

Keywords

  • artificial neural networks
  • geographically weighted regression
  • hedonic regression
  • valuation accuracy

ASJC Scopus subject areas

  • Geography, Planning and Development
  • Urban Studies

Cite this

Prediction accuracy in mass appraisal : A comparison of modern approaches. / Mccluskey, William J.; McCord, M.; Davis, P. T.; Haran, M.; McIlhatton, D.

In: Journal of Property Research, Vol. 30, No. 4, 12.2013, p. 239-265.

Research output: Contribution to journalArticle

Mccluskey, William J. ; McCord, M. ; Davis, P. T. ; Haran, M. ; McIlhatton, D. / Prediction accuracy in mass appraisal : A comparison of modern approaches. In: Journal of Property Research. 2013 ; Vol. 30, No. 4. pp. 239-265.
@article{f7b6123fe16643929b51d401a60b0b77,
title = "Prediction accuracy in mass appraisal: A comparison of modern approaches",
abstract = "The advancement of computational software within the last decade has facilitated enhanced uptake of mass appraisal methodologies by the valuation and prediction accuracy in computer-assisted mass appraisal community for price modelling, estimation and tribunal defence. Applying a sample of 2694 residential properties, this paper assesses and analyses a number of geostatistical approaches relative to an artificial neural network (ANN) model and the traditional linear hedonic pricing model for mass appraisal valuation accuracy and price estimation purposes. The findings demonstrate that the geostatistical localised regression approach is superior in terms of model explanation, reliability and accuracy. ANNs can be shown to perform very well in terms of predictive power, and therefore valuation accuracy, outperforming the traditional multiple regression analysis (MRA) and approaching the performance of spatially weighted regression approaches. However, ANNs retain a 'black box' architecture that limits their usefulness to practitioners in the field. In relation to cost-effectiveness and user-friendly applicability for the valuation community, the MRA approach outperforms the 'black box' nature of the ANN technique, with the geographically weighted regression approach providing the best balance of outright performance and transparency of methodology. It is this spatially weighted approach utilising absolute location which appears to represent the way forward in developing the practice of mass appraisal.",
keywords = "artificial neural networks, geographically weighted regression, hedonic regression, valuation accuracy",
author = "Mccluskey, {William J.} and M. McCord and Davis, {P. T.} and M. Haran and D. McIlhatton",
year = "2013",
month = "12",
doi = "10.1080/09599916.2013.781204",
language = "English",
volume = "30",
pages = "239--265",
journal = "Journal of Property Research",
issn = "0959-9916",
publisher = "Taylor & Francis",
number = "4",

}

TY - JOUR

T1 - Prediction accuracy in mass appraisal

T2 - A comparison of modern approaches

AU - Mccluskey, William J.

AU - McCord, M.

AU - Davis, P. T.

AU - Haran, M.

AU - McIlhatton, D.

PY - 2013/12

Y1 - 2013/12

N2 - The advancement of computational software within the last decade has facilitated enhanced uptake of mass appraisal methodologies by the valuation and prediction accuracy in computer-assisted mass appraisal community for price modelling, estimation and tribunal defence. Applying a sample of 2694 residential properties, this paper assesses and analyses a number of geostatistical approaches relative to an artificial neural network (ANN) model and the traditional linear hedonic pricing model for mass appraisal valuation accuracy and price estimation purposes. The findings demonstrate that the geostatistical localised regression approach is superior in terms of model explanation, reliability and accuracy. ANNs can be shown to perform very well in terms of predictive power, and therefore valuation accuracy, outperforming the traditional multiple regression analysis (MRA) and approaching the performance of spatially weighted regression approaches. However, ANNs retain a 'black box' architecture that limits their usefulness to practitioners in the field. In relation to cost-effectiveness and user-friendly applicability for the valuation community, the MRA approach outperforms the 'black box' nature of the ANN technique, with the geographically weighted regression approach providing the best balance of outright performance and transparency of methodology. It is this spatially weighted approach utilising absolute location which appears to represent the way forward in developing the practice of mass appraisal.

AB - The advancement of computational software within the last decade has facilitated enhanced uptake of mass appraisal methodologies by the valuation and prediction accuracy in computer-assisted mass appraisal community for price modelling, estimation and tribunal defence. Applying a sample of 2694 residential properties, this paper assesses and analyses a number of geostatistical approaches relative to an artificial neural network (ANN) model and the traditional linear hedonic pricing model for mass appraisal valuation accuracy and price estimation purposes. The findings demonstrate that the geostatistical localised regression approach is superior in terms of model explanation, reliability and accuracy. ANNs can be shown to perform very well in terms of predictive power, and therefore valuation accuracy, outperforming the traditional multiple regression analysis (MRA) and approaching the performance of spatially weighted regression approaches. However, ANNs retain a 'black box' architecture that limits their usefulness to practitioners in the field. In relation to cost-effectiveness and user-friendly applicability for the valuation community, the MRA approach outperforms the 'black box' nature of the ANN technique, with the geographically weighted regression approach providing the best balance of outright performance and transparency of methodology. It is this spatially weighted approach utilising absolute location which appears to represent the way forward in developing the practice of mass appraisal.

KW - artificial neural networks

KW - geographically weighted regression

KW - hedonic regression

KW - valuation accuracy

U2 - 10.1080/09599916.2013.781204

DO - 10.1080/09599916.2013.781204

M3 - Article

VL - 30

SP - 239

EP - 265

JO - Journal of Property Research

JF - Journal of Property Research

SN - 0959-9916

IS - 4

ER -