Combining Housing Price Forecasts Generated Separately by Hedonic and Artificial Neural Network Models

Main Article Content

Salvatore Joseph Terregrossa
Mohammed Hussein Ibadi

Abstract

Aims: A) To enhance accuracy in forecasting housing unit prices by forming combinations of component forecasts generated separately by hedonic and artificial neural network models; B) To help ascertain whether a constrained or unconstrained linear combining model achieves superior forecasting performance.

Place and Duration of the Study: Department of Business Administration, Istanbul Aydin University, Istanbul 34295, Turkey; from 2019 to 2020.

Study Design: A cross sectional data set of housing unit prices and corresponding housing unit attributes and characteristics is formed and then randomly divided into two segments: in sample (80%) and out of sample (20%). Three different methods (hedonic, artificial neural network and combining) are then employed to process the same in sample data set, and generate out of sample forecasts. The three forecasting methods are then tested and compared.

Methodology: Out of sample combination forecasts are formed with component forecast weights generated by in sample weighted least squares (WLS) regression of realized price against in sample component forecasts. Four types of regressions are run: unconstrained, with and without a constant; constrained, with and without a constant. Then the mean absolute forecast error of each forecasting method is calculated and the mean difference in absolute forecast error between all pairs of models are compared and tested with a nonparametric Wilcoxon sign rank test.

Results: The combining model formed with component forecast weights generated by weighted least squares (WLS) regression with the constant term suppressed and the sum-of-the-coefficients constrained to equal one, generally performs the best, in comparison with all other forecasting models (component and combination) examined in the study.

Conclusion: The findings represent further evidence regarding the benefits of applying constraints on the linear combining forecast model; and demonstrate that a constrained linear combining model can be a successful technique for enhancing the forecast accuracy of housing unit prices.

Keywords:
Housing price forecasts, hedonic model, artificial neural network model, constrained, linear combining model.

Article Details

How to Cite
Terregrossa, S. J., & Ibadi, M. H. (2021). Combining Housing Price Forecasts Generated Separately by Hedonic and Artificial Neural Network Models. Asian Journal of Economics, Business and Accounting, 21(1), 130-148. https://doi.org/10.9734/ajeba/2021/v21i130345
Section
Original Research Article

References

Terregrossa SJ. Combining analysts' forecasts with causal model forecasts of earnings growth. Applied Financial Economics. 1999; 9:143-153.
DOI: 10.1080/096031099332401.

Bradley M, Gordon J, Mc Manus D. Method for combining house price forecasts. United States Patent. 2003; 660 9109.

Fleming M, Kuo CL. Method and apparatus for predicting and reporting a real estate value based on a weighted average of predicted values. United States Patent. 2007; 7305328.

Drought S, Mc Donald C. Forecasting house price inflation: A model combination approach. Reserve Bank of New Zealand. Discussion Paper Series 32; 2011.

Gupta R, Kabundi A, Miller S. Forecasting the US real house price index: Structural and non-structural models with and without fundamentals. Economic Modelling. 2011; 28(4):2013-2021.

Cabrera J, Wang T, Yang J. Linear and nonlinear predictability of international securitized real estate returns: A reality check. Journal of Real Estate Research. 2011; 33(4):565-594.
DOI: 10.1080/10835547.2011.12091317

Granger CWJ, Ramanathan R. Improved methods of combining forecasts. Journal of Forecasting. 1984; 3: 197-204.
http://dx.doi.org/10.1002/for.3980030207

Clemen RT. Linear-constraints and the efficiency of combined forecasts. Journal of Forecasting. 1986; 5:31-8.
https://doi.org/10.1002/for.3980050104

Terregrossa SJ. On the efficacy of constraints on the linear combination forecast model. Applied Economics Letters. 2005; 12:19-28.
DOI: 10.1080/1350485042000307062.

Guerard JB. Linear constraints, robust weighing and efficient composite modeling. Journal of Forecasting. 1987; 6:193-9.
DOI: doi.org/10.1002/for.3980060305

Lobo G. Alternative methods of combining security analysts’ and statistical forecasts of annual corporate earnings. International Journal of Forecasting. 1991; 7:57-63.
https://doi.org/10.1016/0169-2070(91)90033-R

Griliches Z. ed. Price indexes and quality changes: Studies in new methods of measurement. Cambridge, Mass: Harvard U. Press; 1971.

Rosen S. Hedonic prices and implicit markets. Product differentiation in Pure Competition, J. Political Econ. 1974; 82(1): 34-55.
https://doi.org/10.1086/260169

Calhoun, C A. Property valuation methods and data in the United States. Housing. Finance International. 2001; 16:12-23.

Boardman AE, Greenberg DH, Vining AR, Weiner DL. Cost-benefit analysis: Concepts and practice. 2nd edition. Upper Saddle River, NJ: [Great Britain]: Prentice Hall. 2002; 349-352.

Limsombunchai V, Gan C, Lee M. House price prediction: Hedonic price model vs. artificial neural network. American Journal of Applied Sciences. 2004; 1(3):193-201.
DOI: 10.3844/ajassp.2004.193.201

Lenk MM, Worzala EM, Silva A. High-tech Valuation: Should artificial neural networks bypass the human valuer? J Property Valuation & Investment. 1997; 15:8-26.
DOI: 1108/14635789710163775

Owen C, Howard J. Estimation Realisation Price (ERP) by Neural Networks: Forecasting commercial property values. J Property Valuation & Investment. 1998; 16: 71-86.

Hornik K, Stinchcombe M, White H. Multi-layer feedforward networks are universal approximators. Neural Networks. 1989; 2(5):359-366. https://doi.org/10.1016/0893-6080(89)90020-8

Selim H. Determinants of house prices in Turkey: Hedonic regression versus artificial neural network. Expert Systems with Applications. 2009; 36:2843-2852.
DOI: 10.1016/j.eswa.2008.01.044

Kauko T. On current neural network applications involving spatial modelling of property prices. Journal of Housing and the Built Environment. 2003; 18(2):159-181.
https://doi.org/10.1023/A:1023977111302

Curry B, Morgan P, Silver M. Neural networks and nonlinear statistical methods: An application to the modelling of price quality relationships. Computers & Operations Research. 2002; 29:951-969.
DOI: 10.1016/S0305-0548(00)00096-4

Morano P, Tajani F. Bare ownership evaluation. Hedonic price model vs. artificial neural network. Int. J. Business Intelligence and Data Mining. 2013; 8-4.
DOI: 10.1504/IJBIDM.2013.059263

Stanley M, Alastair A, Dylan M, Patterson D. Neural networks: The prediction of residential values. J Property Valuation & Investment. 1998; 16: 57-70.

Bischoff C W. The combination of macroeconomic forecasts. Journal of Forecasting. 1989; 8: 293-314.
https://doi.org/10.1002/for.3980080312

Wu J, Zhou J, Chen L, Ye L. Coupling forecast methods of multiple rainfall–runoff models for improving the precision of hydrological forecasting. Water Resources Management. 2015; 29: 5091-5108.
DOI: 10.1007/s11269-015-1106-8

Liu X, Moreno B and García S. A grey neural network and input-output combined forecasting model. Primary energy consumption forecasts in Spanish economic sectors. Energy. 2016; 115(1):1042-1054.
DOI: 10.1016/j.energy.2016.09.017

Heng J, Wang J, Xiao L and Lu H. Research and application of a combined model based on frequent pattern growth algorithm and multi-objective optimization for solar radiation forecasting. Applied Energy. 2017; 208:845-866.
DOI: 10.1016/j.apenergy.2017.09.063

Jun W, Yuyan L, Lingyu T, Peng G. Modelling a combined forecast algorithm based on sequence patterns and near characteristics: An application for tourism demand forecasting. Chaos, Solitons & Fractals. Elsevier. 2018; 108:136-147.
DOI: 10.1016/j.chaos.2018.01.028

Wang R, Wang J, Xu Y. A novel combined model based on hybrid optimization algorithm for electrical load forecasting. Applied Soft Computing. 2019; 82:105 548.

Liu Y, Zhang S, Chen X, Wang J. Artificial combined model based on hybrid nonlinear neural network models and statistics linear models-research and application for wind speed forecasting. Sustainability MDPI, Open Access Journal. 2018; 10(12):1-30.
https://doi.org/10.3390/su10124601

Niu X, Wang J. A combined model based on data preprocessing strategy and multi-objective optimization algorithm for short-term wind speed forecasting. Applied Energy. Elsevier. 2019; 241:519–39.
DOI: 10.1016/j.apenergy.2019.03.097

Liu Z, Jiang P, Zhang L, Niu X. A combined forecasting model for time series: Application to short-term wind speed forecasting. Applied Energy Elsevier. 2020; 259:1-25.
DOI: 10.1016/j.apenergy.2019.114137

Istanbul real estate.
Available: www.istanbulrealestate.com.
Accessed: 15th, 16th and 17th May 2019.

Istanbul Property World.
Available:www.istanbulpropertyworld.com.
Accessed: 19th to 22nd May 2019].

Property Turkey.
Available:www.propertyturkey.com.
Accessed: 29th May to 10th June 2019.

Turkey Expert.
Available:www.turkeyexpert.com.
Accessed: 20th, 21st&30th June 2019.

Istanbul Homes.
Available:www.istanbulhomes.com.
Accessed from 3rd to 6th June 2019.

Turkey Homes.
Available: www.turkeyhomes.com.
Accessed: 1st to 9th July 2019.

Halvorsen R, Palmquist R. The interpretation of dummy variables in semi-logarithmic equations. American Economic Review. 1980; 70(3):474-475.

Goodman A, Thibodeau T. Dwelling age heteroscedasticity in hedonic house price equations. J Housing Res. 1995; 6:25-42.

Goodman A, Thibodeau T. Dwelling age heteroscedasticity in hedonic house price equations: An extension. J Housing Res. 1997; 8:299-317.

Stevenson S. New empirical evidence on heteroscedasticity in hedonic housing models. Journal of Housing Economics. 2004; 13:136-153.
DOI: 10.1016/j.jhe.2004.04.004

Fletcher M, Gallimore P, Mangan J. Heteroscedasticity in hedonic house price models. J Property Res. 2000a; 17:93-108.
DOI: 10.1080/095999100367930