Abstract
We present a Bayesian hedonic regression model for the appraisal of apartments located within the metropolitan area of Belgrade, Serbia. Data on 12.904 apartments were collected from a local classified website, and used to fit and validate the model. The response variable, which is the price per m\(^2\) in euros, is assumed to be log-normally distributed. Nested random effects are used to model the hierarchical structure present in the location identifiers, and thin-plate spline functions are used to capture the nonlinear effects. Location explains around 78.62% of the total variation in advertised prices per m\(^2\) in the training set, confirming its paramount importance. Further major factors affecting prices are: area in m\(^2\), floor number, the total number of floors in the building, the availability of an elevator and condition. The predictive functionality of the model is demonstrated through an open access online application that accompanies this study. The model achieves Mean Average Percent Error (MAPE) of 13.06% in the validation set. Its predictive performance is compared to that of three popular machine learning (ML) methods, and its suitability for mass appraisal is examined. This is the first empirical study to present the results obtained from a hedonic regression built using data pertaining to Belgrade.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
Notes
- 1.
In Serbia there is an available organized database of transactional property prices, maintained by The Republic Geodetic Authority (https://katastar.rgz.gov.rs/RegistarCenaNepokretnosti/). However, the structure of data does not allow for an adequate assessment of the predictive accuracy of the models presented in this paper; except for the address, the date of transaction, the size and the cadastral parcel, any other information about the quality and the state of the property is missing. The database is available both for public and professional access, but there is a very small difference in the quality of information obtained in both. However, professional access is costly for any sort of research which demands a large amount of data about transactional prices.
- 2.
Asking price is “a price at which a seller commits to taking his good off the market and trading immediately” as defined in [35].
- 3.
Market price is “the estimated amount for which the property should exchange on the date of valuation between a willing buyer and a willing seller in an arm’s-length transaction after proper marketing wherein the parties had each acted knowledgeably, prudently and without being under compulsion”, as defined in [36].
References
Kang, H.-B., Reichert, A.K.: An empirical analysis of Hedonic regression and grid-adjustment techniques in real estate appraisal. Real Estate Econ. 19, 70–91 (1991).https://doi.org/10.1111/1540-6229.00541
Keskin, B.: Hedonic analysis of price in the Istanbul housing market. Int. J. Strateg. Prop. Manag. 12, 125–138 (2008). https://doi.org/10.3846/1648-715X.2008.12.125-138
Wang, D., Li, V.J.: Mass appraisal models of real estate in the 21st century: a systematic literature review. Sustainability 11, 7006 (2019). https://doi.org/10.3390/su11247006
Court, A.T.: Hedonic price indexes with automotive examples. In: Ross, C.F. (ed.) The Dynamics of Automotive Demand, pp. 99–117. General Motors, New York (1939)
Wallace, H.A.: Comparative farm-land values in Iowa. J. L. Public Util. Econ. 2, 385 (1926). https://doi.org/10.2307/3138610
Haas, G.: A statistical analysis of farm sales in Blue Earth County, Minnesota, as a basis for Farm Land Appraisal (1922)
Griliches, Z.: Notes on the measurement of price and quality changes. In: Models of Income Determination, pp. 381–418. National Bureau of Economic Research, Inc., Cambridge (1964)
Herath, S., Maier, G.: The hedonic price method in real estate and housing market research. A review of the literature, Vienna (2010)
Peirson, A., Jovanovic, J.: Belgrade Residential Market | Snapshot Q4 2019, Belgrade (2020)
Bajat, B., Kilibarda, M., Pejović, M., Petrović, M.S.: Spatial hedonic modeling of housing prices using auxiliary maps. In: Thill, J.-C. (ed.) Advances in Geographic Information Science, pp. 97–122. Springer, Berlin, Heidelberg (2018). https://doi.org/10.1007/978-3-642-37896-6_5
Vidović, M., Radosavljević, I., Mitrović, A., Konjović, Z.: Construction of Training Data for price prediction of a real estate from internet ads. In: Proceedings of the International Scientific Conference - Sinteza 2019, pp. 388–393. Singidunum University, Novi Sad, Serbia (2019). https://doi.org/10.15308/Sinteza-2019-388-393
Browne, W.J., Goldstein, H., Rasbash, J.: Multiple membership multiple classification (MMMC) models. Stat. Modell. 1, 103–124 (2001). https://doi.org/10.1191/147108201128113
R Core Team.: R: a language and environment for statistical computing. http://www.r-project.org (2019)
Bürkner, P.-C.: brms: an R package for Bayesian multilevel models using Stan. J. Stat. Softw. 80, 1–28 (2017). https://doi.org/10.18637/jss.v080.i01
Gelman, A., Goodrich, B., Gabry, J., Vehtari, A.: R-squared for Bayesian regression models. Am. Stat. 73, 307–309 (2019). https://doi.org/10.1080/00031305.2018.1549100
Valier, A.: Who performs better? AVMs vs hedonic models. J. Prop. Invest. Financ. (2020). https://doi.org/10.1108/JPIF-12-2019-0157
Kontrimas, V., Verikas, A.: The mass appraisal of the real estate by computational intelligence. Appl. Soft Comput. 11, 443–448 (2011). https://doi.org/10.1016/J.ASOC.2009.12.003
Peterson, S., Flanagan, A.: Neural network hedonic pricing models in mass real estate appraisal. J. Real Estate Res. 31, 147–164 (2009). https://doi.org/10.5555/rees.31.2.m3h27210w6411373
Antipov, E.A., Pokryshevskaya, E.B.: Mass appraisal of residential apartments: an application of Random forest for valuation and a CART-based approach for model diagnostics. Expert Syst. Appl. 39, 1772–1778 (2012). https://doi.org/10.1016/j.eswa.2011.08.077
Zurada, J., Levitan, A., Guan, J.: A comparison of regression and artificial intelligence methods in a mass appraisal context. J. Real Estate Res. 33, 349–387 (2011). https://doi.org/10.5555/rees.33.3.q6890722u7375871
McCluskey, W.J., McCord, M., Davis, P.T., Haran, M., McIlhatton, D.: Prediction accuracy in mass appraisal: a comparison of modern approaches. J. Prop. Res. 30, 239–265 (2013). https://doi.org/10.1080/09599916.2013.781204
Kuhn, M.: caret: classification and regression training (2017). https://cran.r-project.org/package=caret
Fik, T.J., Ling, D.C., Mulligan, G.F.: Modeling spatial variation in housing prices: a variable interaction approach. Real Estate Econ. 31, 623–646 (2003). https://doi.org/10.1046/j.1080-8620.2003.00079.x
Christodoulou, E., Ma, J., Collins, G.S., Steyerberg, E.W., Verbakel, J.Y., Van Calster, B.: A systematic review shows no performance benefit of machine learning over logistic regression for clinical prediction models. J. Clin. Epidemiol. 110, 12–22 (2019). https://doi.org/10.1016/J.JCLINEPI.2019.02.004
Radojevic, T., Stanisic, N., Stanic, N.: The culture of hospitality: From anecdote to evidence. Ann. Tour. Res. 79, 102789 (2019). https://doi.org/10.1016/j.annals.2019.102789
Radojevic, T., Stanisic, N., Stanic, N.: Inside the rating scores: a multilevel analysis of the factors influencing customer satisfaction in the hotel industry. Cornell Hosp. Q. 58 (2017). https://doi.org/10.1177/1938965516686114
Radojevic, T., Stanisic, N., Stanic, N.: Ensuring positive feedback: factors that influence customer satisfaction in the contemporary hospitality industry. Tour. Manag. 51, 13–21 (2015). https://doi.org/10.1016/j.tourman.2015.04.002
Radojevic, T., Stanisic, N., Stanic, N., Davidson, R.: The effects of traveling for business on customer satisfaction with hotel services. Tour. Manag. 67, 326–341 (2018). https://doi.org/10.1016/j.tourman.2018.02.007
Selim, H.: Determinants of house prices in Turkey: Hedonic regression versus artificial neural network. Expert Syst. Appl. 36, 284–852 (2009). https://doi.org/10.1016/J.ESWA.2008.01.044
Hui, S.K., Cheung, A., Pang, J.: A hierarchical bayesian approach for residential property valuation: application to Hong Kong housing market. Int. Real Estate Rev. 13, 1–29 (2010)
Brown, K., Uyar, B.: A hierarchical linear model approach for assessing the effects of house and neighborhood characteristics on housing prices. J. Real Estate Pract. Educ. 7, 15–24 (2004). https://doi.org/10.5555/repe.7.1.f687057161743261
Arribas, I., García, F., Guijarro, F., Oliver, J., Tamošiūnienė, R.: Mass appraisal of residential real estate using multilevel modelling. Int. J. Strateg. Prop. Manag. 20, 77–87 (2016). https://doi.org/10.3846/1648715X.2015.1134702
Deboosere, R., Kerrigan, D.J., Wachsmuth, D., El-Geneidy, A.: Location, location and professionalization: a multilevel hedonic analysis of Airbnb listing prices and revenue. Reg. Stud. Reg. Sci. 6, 143–156 (2019). https://doi.org/10.1080/21681376.2019.1592699
Bao, H.X.H., Wan, A.T.K.: On the use of spline smoothing in estimating hedonic housing price models: empirical evidence using Hong Kong data. Real Estate Econ. 32, 487–507 (2004). https://doi.org/10.1111/j.1080-8620.2004.00100.x
Lester, B., Visschers, L., Wolthoff, R.: Competing with asking prices. Theor. Econ. 12, 731–770 (2017). https://doi.org/10.3982/te1846
Regulation (EU): EUR-Lex-32013R0575-EN-EUR-Lex (2013)
Acknowledgements
This research was supported by the Science Fund of the Republic of Serbia, #GRANT No. 65241005, AI—ATLAS.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Appendix
Appendix
Appendix A: Estimated Regression Coefficients
Population-level effect | Estimate | Estimate Error | l-95 % CI | u-95 % CI | Exponentiated coefficient estimate |
|---|---|---|---|---|---|
Intercept | 7.12 | 0.23 | 6.67 | 7.59 | 1236.45 |
s(Area_in_sqm) | −1.60 | 0.84 | −3.21 | 0.00 | 0.20 |
s(Num_of_rooms) | −0.01 | 0.08 | −0.18 | 0.15 | 0.99 |
\(\text {s(Floor\_num} \times \text {Stories)}\) | −2.13 | 0.36 | −2.82 | −1.45 | 0.12 |
\(\text {s(Floor\_num} \times \text {Stories)2}\) | 0.79 | 0.33 | 0.14 | 1.43 | 2.20 |
\(\text {s(Floor\_num} \times \text {Elevator[No])}\) | −1.57 | 47.90 | −88.73 | 113.03 | 0.21 |
Elevator[No] | −0.19 | 0.23 | −0.57 | 0.34 | 0.83 |
Not_at_top_floor[No] | −0.02 | 0.00 | −0.02 | −0.01 | 0.98 |
Registered_in_cadaster [No] | −0.02 | 0.00 | −0.03 | −0.01 | 0.98 |
Move_in_ready[No] | 0.01 | 0.00 | 0.00 | 0.01 | 1.01 |
VAT_refund[No] | 0.00 | 0.01 | −0.01 | 0.01 | 1.00 |
Belgrade_WF[Yes] | 0.28 | 0.17 | −0.05 | 0.60 | 1.32 |
Duplex[Yes] | 0.07 | 0.06 | −0.05 | 0.18 | 1.07 |
\(\text {Belgrade\_WF[Yes]} \times \text {Duplex[Yes]}\) | 0.16 | 0.08 | 0.00 | 0.32 | 1.17 |
Balcony[No] | −0.02 | 0.00 | −0.02 | −0.01 | 0.98 |
Recessed_balcony[No] | 0.00 | 0.00 | −0.01 | 0.00 | 1.00 |
Phone[No] | 0.01 | 0.00 | 0.00 | 0.01 | 1.01 |
Intercom[No] | −0.01 | 0.00 | −0.02 | 0.00 | 0.99 |
Air_conditioner[No] | −0.01 | 0.00 | −0.02 | −0.01 | 0.99 |
Surveillance[No] | −0.01 | 0.00 | -0.02 | -0.01 | 0.99 |
Internet[No] | 0.01 | 0.00 | 0.00 | 0.01 | 1.01 |
Cable_TV[No] | 0.00 | 0.00 | 0.00 | 0.01 | 1.00 |
Garage[No] | −0.05 | 0.00 | −0.06 | −0.04 | 0.95 |
Parking[No] | 0.01 | 0.00 | 0.00 | 0.01 | 1.01 |
Basement_storage[No] | −0.01 | 0.00 | −0.01 | 0.00 | 0.99 |
Condition[Lux] | 0.01 | 0.00 | 0.00 | 0.01 | 1.01 |
Condition[Renovated] | −0.07 | 0.00 | −0.08 | −0.06 | 0.93 |
Condition[MissingInfo] | 0.05 | 0.00 | 0.04 | 0.06 | 1.05 |
Type[NewBuilt] | 0.01 | 0.00 | 0.00 | 0.02 | 1.01 |
Type[UnderConstruction] | −0.08 | 0.01 | −0.09 | −0.07 | 0.92 |
Type[MissingInfo] | −0.01 | 0.01 | −0.03 | 0.00 | 0.99 |
Heating[DistrictHeating] | 0.02 | 0.00 | 0.01 | 0.03 | 1.02 |
Heating[ElectricThermalStorage] | −0.06 | 0.01 | −0.07 | −0.05 | 0.94 |
Heating[MarbleHeaters] | −0.03 | 0.03 | −0.10 | 0.02 | 0.97 |
Heating[MissingInfo] | 0.03 | 0.01 | 0.01 | 0.04 | 1.03 |
Heating[NaturalGas] | 0.02 | 0.01 | 0.00 | 0.04 | 1.02 |
Heating[NorwegianPanelHeaters] | −0.05 | 0.01 | −0.08 | −0.02 | 0.95 |
Heating[TiledStove] | −0.04 | 0.03 | −0.11 | 0.02 | 0.96 |
Heating[UnderfloorHeating] | 0.06 | 0.02 | 0.02 | 0.09 | 1.06 |
Advertiser[Bank] | −0.02 | 0.12 | −0.26 | 0.21 | 0.98 |
Advertiser[Investor] | −0.02 | 0.02 | −0.06 | 0.02 | 0.98 |
Advertiser[Owner] | 0.00 | 0.01 | −0.02 | 0.02 | 1.00 |
AgencyFee[1%] | 0.06 | 0.09 | −0.11 | 0.23 | 1.06 |
AgencyFee[1.50%] | 0.05 | 0.15 | −0.26 | 0.35 | 1.05 |
AgencyFee[2%] | 0.03 | 0.02 | −0.01 | 0.08 | 1.03 |
AgencyFee[2.40%] | 0.15 | 0.10 | −0.06 | 0.36 | 1.16 |
AgencyFee[MissingInfo] | 0.02 | 0.02 | −0.02 | 0.07 | 1.02 |
Rights and permissions
Copyright information
© 2021 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this chapter
Cite this chapter
Stanišić, N., Radojević, T., Stanić, N. (2021). Appraisal of Apartments in Belgrade Using Hedonic Regression: Model Specification, Predictive Performance, Suitability for Mass Appraisal, and Comparison with Machine Learning Methods. In: Pap, E. (eds) Artificial Intelligence: Theory and Applications. Studies in Computational Intelligence, vol 973. Springer, Cham. https://doi.org/10.1007/978-3-030-72711-6_16
Download citation
DOI: https://doi.org/10.1007/978-3-030-72711-6_16
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-72710-9
Online ISBN: 978-3-030-72711-6
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)