Integrating Machine Learning and Hedonic Regression for Housing Price Prediction: A Systematic International Review of Model Performance and Interpretability
Mahshid Gorjian
MPRA Paper from University Library of Munich, Germany
Abstract:
It is becoming increasingly important to predict property prices to mitigate investment risk, establish policies, and preserve market stability. To determine the practical utility and anticipated efficacy of the sophisticated statistical and machine learning models that have emerged, a comparative analysis is required. The purpose of this systematic study is to assess the predictive effectiveness and interpretability of hedonic regression and complex machine learning models in the estimation of housing prices in a wide range of foreign scenarios. In May 2024, a thorough search was conducted in Scopus, Google Scholar, and Web of Science. The search terms included "hedonic pricing models," "machine learning," and "housing price prediction," in addition to others. The inclusion criteria required the utilization of empirical research published after 2000, a comparison of at least two predictive models, and reliable transaction data. Research that utilized non-empirical methodologies or web- scraped prices was excluded. Twenty-three investigations met the eligibility criteria. The evaluation was conducted in accordance with the reporting criteria of PRISMA 2020. Random Forest was the most frequently employed and consistently high-performing model, being selected in 14 of 23 studies and regarded as exceptional in five. Despite their lack of precision, hedonic regression models provided critical explanatory insights into critical variables, such as proximity to urban centers, property characteristics, and location. The integration of hedonic and machine learning models improved the interpretability and accuracy of the predicted results. Many of the studies included in this review were longitudinal, covered a diverse range of international contexts (specifically, Asia, Europe, America, and Australia), and demonstrated a rise in research output beyond 2020. Even though hedonic models retain a significant amount of explanatory power, the precision of home price predictions is improved by machine learning, particularly Random Forest and neural networks. The optimal results for researchers, real estate professionals, and policymakers who aim to improve market transparency and enlighten effective policy decisions are achieved through the seamless integration of these techniques.
Keywords: housing price prediction; machine learning; hedonic price model; Random Forest; real estate valuation; artificial neural networks; systematic review; property market analysis (search for similar items in EconPapers)
JEL-codes: C00 C01 C10 (search for similar items in EconPapers)
Date: 2025
New Economics Papers: this item is included in nep-big and nep-cmp
References: View references in EconPapers View complete reference list from CitEc
Citations:
Downloads: (external link)
https://mpra.ub.uni-muenchen.de/125676/1/MPRA_paper_125676.pdf original version (application/pdf)
Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.
Export reference: BibTeX
RIS (EndNote, ProCite, RefMan)
HTML/Text
Persistent link: https://EconPapers.repec.org/RePEc:pra:mprapa:125676
Access Statistics for this paper
More papers in MPRA Paper from University Library of Munich, Germany Ludwigstraße 33, D-80539 Munich, Germany. Contact information at EDIRC.
Bibliographic data for series maintained by Joachim Winter ().