A Comparative Analysis of Hedonic OLS and Random Forest Models for Apartment Price Estimation
Downloads
Abstract
This study investigates apartment price formation through a comparative assessment of econometric and Machine Learning (ML) models applied to a micro-level dataset of residential properties in Tirana, Albania. The research addresses the challenge of modelling housing prices in emerging real estate markets characterized by heterogeneous property attributes and spatial variation. To ensure comparability across dwellings of different sizes, the dependent variable is defined as price per square meter. The methodological framework combines a hedonic Ordinary Least Squares (OLS) model and a nonlinear Random Forest (RF) model, allowing the evaluation of both model interpretability and predictive performance. Elastic Net and Extreme Gradient Boosting (XGBoost) models are additionally employed as robustness benchmarks. Model performance is assessed using standard prediction accuracy measures, while variable effects and importance metrics are analysed to identify the main determinants of housing prices. The results reveal that location-related factors and structural housing characteristics constitute the dominant drivers of apartment values. Apartment size is negatively associated with price per square meter, whereas the number of bathrooms is positively and statistically significantly associated. The number of rooms becomes insignificant after controlling for other explanatory variables. Strong neighbourhood effects confirm substantial spatial heterogeneity in the housing market. The comparative analysis demonstrates that RF achieves superior predictive accuracy relative to the alternative models, highlighting the ability of nonlinear methods to capture complex relationships in housing price data. The findings contribute to the application of statistical and ML techniques in real estate valuation and provide evidence on the relative strengths of interpretable and predictive modelling approaches.
Keywords
Housing price modelling Hedonic regression Random forest Real estate valuation Machine learning Spatial heterogeneityReferences (41)
- [1] Zhao, C., & Liu, F. (2023). Impact of housing policies on the real estate market-systematic literature review. Heliyon, 9(10), e20704. https://www.cell.com/heliyon/fulltext/S2405-8440(23)07912-4
- [2] Han, F., Lu, M., Qin, D., Zheng, G., Zeng, G., Tan, Y., ... & He, H. (2025). Exploring housing price dynamics in sustainable cities through a cooperated big data driven machine learning method: Case study on a typical city in China. City and environment interactions, 28, 100223. https://doi.org/10.1016/j.cacint.2025.100223
- [3] Çılgın, C., & Gökçen, H. (2023). Machine learning methods for prediction real estate sales prices in Turkey. Revista de la construcción, 22(1), 163-177. http://dx.doi.org/10.7764/rdlc.22.1.163
- [4] Pojani, D. (2010). Tirana. Cities, 27(6), 483–495. https://doi.org/10.1016/j.cities.2010.02.002
- [5] Thanasi, M. (2016). Hedonic appraisal of apartments in Tirana. International journal of housing markets and analysis, 9(2), 239–255. https://doi.org/10.1108/IJHMA-03-2015-0016
- [6] Liu, T., Wang, J., Liu, L., Peng, Z., & Wu, H. (2025). What are the pivotal factors influencing housing prices? A spatiotemporal dynamic analysis across market cycles from upturn to downturn in Wuhan. Land, 14(2), 356. https://doi.org/10.3390/land14020356
- [7] Chwiałkowski, C., Zydroń, A., & Kayzer, D. (2022). Assessing the impact of selected attributes on dwelling prices using ordinary least squares regression and geographically weighted regression: A case study in Poznań, Poland. Land, 12(1), 125. https://doi.org/10.3390/land12010125
- [8] Shtepani, E., & Yunitsyna, A. (2023). Evaluation of the spatial quality of apartments from different price categories using the visibility graph analysis: A case of Tirana, Albania. International journal of real estate studies, 17(1), 83–92. https://doi.org/10.11113/intrest.v17n1.268
- [9] Nurja, I., Jaupi, F., & Elezaj, O. (2022). Appraisal of apartments in Albania using hedonic regression. WSEAS transactions on business and economics, 19, 1816–1823. https://doi.org/10.37394/23207.2022.19.163
- [10] Ko, D., & Park, S. (2024). Investigating the correlation between air pollution and housing prices in Seoul, South Korea: Application of explainable artificial intelligence in random forest machine learning. Sustainability, 16(11), 4453. https://doi.org/10.3390/su16114453
- [11] Kim, W., & Hong, J. (2024). Stacked ensemble model for the automatic valuation of residential properties in South Korea: A case study on Jeju Island. Land, 13(9), 1436. https://doi.org/10.3390/land13091436
- [12] Wan, H., Chowdhury, P. K. R., Yoon, J., Bhaduri, P., Srikrishnan, V., Judi, D., & Daniel, B. (2025). Explaining drivers of housing prices with nonlinear hedonic regressions. Machine learning with applications, 21, 100707. https://doi.org/10.1016/j.mlwa.2025.100707
- [13] Wei, C., Fu, M., Wang, L., Yang, H., Tang, F., & Xiong, Y. (2022). The research development of hedonic price model-based real estate appraisal in the era of big data. Land, 11(3), 334. https://doi.org/10.3390/land11030334
- [14] Khoshnoud, M., Sirmans, G. S., & Zietz, E. N. (2023). The evolution of hedonic pricing models. Journal of real estate literature, 31(1), 1–47. https://doi.org/10.1080/09277544.2023.2201020
- [15] Rey-Blanco, D., Zofio, J. L., & Gonzalez-Arias, J. (2024). Improving hedonic housing price models by integrating optimal accessibility indices into regression and random forest analyses. Expert systems with applications, 235, 121059. https://doi.org/10.1016/j.eswa.2023.121059
- [16] Marinković, S., Džunić, M., & Marjanović, I. (2024). Determinants of housing prices: Serbian cities’ perspective. Journal of housing and the built environment, 39(3), 1601–1626. https://doi.org/10.1007/s10901-024-10134-5
- [17] Shehu, E., Afezolli, A., & Kondi, I. (2015). The model for determining the market value of residential properties in Tirana city. Proceedings of international conference on innovation in civil and structural engineering (ICICSE) (pp. 164-169). Unique Conferences Publishing. http://dx.doi.org/10.17758/UR.U0615307
- [18] Wang, Z., Wang, Y., Xia, X., Chen, S., & Jiang, W. (2025). How does built environment influence housing prices in large-scale areas? An interpretable machine learning method by considering multi-dimensional accessibility. ISPRS international journal of Geo-information, 14(11), 436. https://doi.org/10.3390/ijgi14110436
- [19] Zhang, Y., & Miller, E. J. (2025). Location choice of residential housing supply: An application of the multiple discrete-continuous extreme value (MDCEV) model. Journal of choice modelling, 54, 100535. https://doi.org/10.1016/j.jocm.2024.100535
- [20] Dou, M., Gu, Y., & Fan, H. (2023). Incorporating neighborhoods with explainable artificial intelligence for modeling fine-scale housing prices. Applied geography, 158, 103032. https://doi.org/10.1016/j.habitatint.2024.103212
- [21] Li, J., Ossokina, I. V, & Arentze, T. A. (2024). The impact of urban green space on housing value: A combined hedonic price analysis and land use modeling approach. Journal of sustainable real estate, 16(1), 2432758. https://doi.org/10.1080/19498276.2024.2432758
- [22] Li, C., Zhou, Y., Wu, M., Xu, J., & Fu, X. (2025). Exploring nonlinear threshold effects and interactions between built environment and urban vitality at the block level using machine learning. Land, 14(6), 1232. https://doi.org/10.3390/land14061232
- [23] Anelli, D., Morano, P., Tajani, F., & Guarini, M. R. (2025). The interpretative effects of normalization techniques on complex regression modeling: An application to real estate values using machine learning. Information, 16(6), 486. https://doi.org/10.3390/info16060486
- [24] Choy, L. H. T., & Ho, W. K. O. (2023). The use of machine learning in real estate research. Land, 12(4), 740. https://doi.org/10.3390/land12040740
- [25] Moreno-Foronda, I., Sánchez-Martínez, M. T., & Pareja-Eastaway, M. (2025). Comparative analysis of advanced models for predicting housing prices: A review. Urban science, 9(2), 32. https://doi.org/10.3390/urbansci9020032
- [26] Maselli, G., & Nesticò, A. (2025). Machine learning algorithms and explainable artificial intelligence for property valuation. Real estate, 2(3), 12. https://doi.org/10.3390/realestate2030012
- [27] Soltani, A., & Lee, C. L. (2024). The non-linear dynamics of South Australian regional housing markets: A machine learning approach. Applied geography, 166, 103248. https://doi.org/10.1016/j.apgeog.2024.103248
- [28] Jafary, P., Shojaei, D., Rajabifard, A., & Ngo, T. (2025). AI, machine learning and BIM for enhanced property valuation: Integration of cost and market approaches through a hybrid model. Habitat international, 164, 103515. https://doi.org/10.1016/j.habitatint.2025.103515
- [29] Andrade-Girón, D. C., Marin-Rodriguez, W. J., & Zuñiga-Rojas, M. G. (2025). Intelligent feature selection ensemble model for price prediction in real estate markets. Informatics, 12(2), 52. https://doi.org/10.3390/informatics12020052
- [30] Yang, Y., & Wang, H. (2025). Random forest-based machine failure prediction: A performance comparison. Applied sciences, 15(16), 8841. https://doi.org/10.3390/app15168841
- [31] Zhang, Q., & Abdullah, F. (2026). Hedonic beats utilitarian: Differential effects of AI chatbots and AR/VR on consumer engagement in e-commerce. Journal of theoretical and applied electronic commerce research, 21(2), 60. https://doi.org/10.3390/jtaer21020060
- [32] Pugliese, R., Regondi, S., & Marini, R. (2021). Machine learning-based approach: Global trends, research directions, and regulatory standpoints. Data science and management, 4, 19–29. https://doi.org/10.1016/j.dsm.2021.12.002
- [33] An, S., Song, Y., Jang, H., & Ahn, K. (2025). Toward transparent and accurate housing price appraisal: Hedonic price models versus machine learning algorithms. Financial innovation, 11(1), 141. https://doi.org/10.1186/s40854-025-00874-w
- [34] Rosen, S. (1974). Hedonic prices and implicit markets: Product differentiation in pure competition. Journal of political economy, 82(1), 34–55. https://doi.org/10.1086/260169
- [35] Breiman, L. (2001). Random forests. Machine learning, 45(1), 5–32. https://doi.org/10.1023/A:1010933404324
- [36] James, G. (2013). An introduction to statistical learning with applications in R. Springer. https://doi.org/10.1007/978-3-031-38747-0
- [37] Kuhn, M. (2013). Applied predictive modeling. Springer. https://doi.org/10.1007/978-1-4614-6849-3
- [38] O’brien, R. M. (2007). A caution regarding rules of thumb for variance inflation factors. Quality & quantity, 41(5), 673–690. https://doi.org/10.1007/s11135-006-9018-6
- [39] Breusch, T. S., & Pagan, A. R. (1979). A simple test for heteroscedasticity and random coefficient variation. Econometrica: Journal of the econometric society, 47(5), 1287–1294. https://doi.org/10.2307/1911963
- [40] White, H. (1980). A heteroskedasticity-consistent covariance matrix estimator and a direct test for heteroskedasticity. Econometrica: Journal of the econometric society, 48(4), 817–838. https://doi.org/10.2307/1912934
- [41] Zou, H., & Hastie, T. (2005). Regularization and variable selection via the elastic net. Journal of the royal statistical society series b: Statistical methodology, 67(2), 301–320. https://doi.org/10.1111/j.1467-9868.2005.00527.x
