Comparative evaluation of the performance of different regression models in land valuation
Article Sidebar
Main Article Content
Abstract
Lands can play a dominant role in the real estate market, especially due to their legal zoning rights. These properties are preferred investment options compared to financial instruments due to factors such as high returns and long-term reliability. Today, Machine Learning (ML) algorithms are used to accurately determine the land value. Regression models, capable of handling complex relationships, integrating Geographic Information System (GIS), and providing a comparative approach, lead the way among these algorithms. In this study, Lasso, Elastic-Net, ML.Net, and Ordinary Least Squares (OLS) regression models were employed to predict land values in the central neighborhoods of Konya's Selçuklu, Meram, and Karatay districts. The datasets containing legal, physical, spatial, and local criteria of 440 lands were obtained, and GIS analyses were conducted to prepare the spatial data. Based on the modeling results, it can be observed that ML.Net exhibited successful performance with metric values of MAE=0.043, MSE=0.005, RMSE=0.060, and R2=0.82. Comparatively, ML.Net's 9% superior performance compared to the commonly encountered OLS in the literature is of significant importance. The results demonstrated the usability of various regression models for land valuation and highlighted that ML.Net can yield improved outcomes, particularly in modeling high-market-value lands.
Article Details
References
Demetriou, D. (2016). The assessment of land valuation in land consolidation schemes: The need for a new land valuation framework. Land use policy, 54, 487-498. https://doi.org/10.1016/j.landusepol.2016.03.008
Derdouri, A., & Murayama, Y. (2020). A comparative study of land price estimation and mapping using regression kriging and machine learning algorithms across Fukushima prefecture, Japan. Journal of Geographical Sciences, 30, 794-822. https://doi.org/10.1007/s11442-020-1756-1
Dismuke, C., & Lindrooth, R. (2006). Ordinary least squares. Methods and designs for outcomes research, 93(1), 93-104.
Doan, Q. C. (2023). Determining the optimal land valuation model: A case study of Hanoi, Vietnam. Land use policy, 127, 106578. https://doi.org/10.1016/j.landusepol.2023.106578
Foryś, I., & Gaca, R. (2018). Intuitive methods versus analytical methods in real estate valuation: preferences of Polish real estate appraisers. In Problems, Methods and Tools in Experimental and Behavioral Economics: Computational Methods in Experimental Economics (CMEE) 2017 Conference. Łódź, Poland 79-87
Hu, S., Yang, S., Li, W., Zhang, C., & Xu, F. (2016). Spatially non-stationary relationships between urban residential land price and impact factors in Wuhan city, China. Applied Geography, 68, 48-56. https://doi.org/10.1016/j.apgeog.2016.01.006
Kokot, S., & Gnat, S. (2019). Simulative verification of the possibility of using multiple regression models for real estate appraisal. Real Estate Management and Valuation, 27(3), 109-123. https://doi.org/10.2478/remav-2019-0029
Krause, A. L., & Bitter, C. (2012). Spatial econometrics, land values and sustainability: Trends in real estate valuation research. Cities, 29, 19S25. https://doi.org/10.1016/j.cities.2012.06.006
Microsoft. (2018). ML.Net: Machine Learning made for .Net. Microsoft.
Ramel, D. (2018). Open Source, Cross-Platform ML.Net Simplifies Machine Learning - Visual Studio Magazine. Visual Studio Magazine.
Sisman, S. & Aydinoglu, A. C. (2022). Improving performance of mass real estate valuation through application of the dataset optimization and Spatially Constrained Multivariate Clustering Analysis. Land use policy, 119, 106167. https://doi.org/10.1016/j.landusepol.2022.106167
Sisman, S., Akar, A. U., & Yalpir, S. (2023). The novelty hybrid model development proposal for mass appraisal of real estates in sustainable land management. Survey Review, 55(388), 1-20. https://doi.org/10.1080/00396265.2021.1996797
Tibshirani, R. (1996). Regression shrinkage and selection via the lasso. Journal of the Royal Stat. Society Series B: Statistical Methodology, 58(1), 267-288. https://doi.org/10.1111/j.2517-6161.1996.tb02080.x
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.1467-9868.2005.00503.x
Zurada, J., Levitan, A., & Guan, J. (2011). A comparison of regression and artificial intelligence methods in a mass appraisal context. Journal of real estate research, 33(3), 349-388. https://doi.org/10.1080/10835547.2011.12091311