Interpretable machine learning for real estate market analysis
Интерпретируемое машинное обучение для анализа рынка недвижимости
2022-05-31
SCID: 54.1/hty5ncfk
Discuss with AI
distance decay functionshedonic machine learninginterpretable machine learningmodel-agnostic interpretationreal estate market analysis
Figures from the paper
Abstract (AI)
Abstract Machine Learning (ML) excels at most predictive tasks but its complex nonparametric structure renders it less useful for inference and out‐of sample predictions. This article aims to elucidate and enhance the analytical capabilities of ML in real estate through Interpretable ML (IML). Specifically, we compare a hedonic ML approach to a set of model‐agnostic interpretation methods. Our results suggest that IML methods permit a peek into the black box of algorithmic decision making by showing the web of associative relationships between variables in greater resolution. In our empirical applications, we confirm that size and age are the most important rent drivers. Further analysis reveals that certain bundles of hedonic characteristics, such as large apartments in historic buildings with balconies located in affluent neighborhoods, attract higher rents than adding up the contributions of each hedonic characteristic. Building age is shown to exhibit a U‐shaped pattern in that both the youngest and oldest buildings attract the highest rents. Besides revealing valuable distance decay functions for spatial variables, IML methods are also able to visualise how the strength and interactions of hedonic characteristics change over time, which investors could use to determine the types of assets that perform best at any given stage of the real estate investment cycle.
Key Findings
1
Apartment size and building age are identified as the most important drivers of rents in the empirical applications.
2
Building age has a U-shaped relationship with rent: the youngest and oldest buildings attract the highest rents.
3
Certain characteristic bundles—large apartments in historic buildings with balconies in affluent neighborhoods—command higher rents than the summed individual effects predict.
4
Interpretable methods expose spatial distance-decay functions and changing hedonic effects and interactions over time, supporting investment-cycle asset selection.
5
Model-agnostic interpretation methods reveal associative relationships among housing variables in greater detail than black-box predictions alone.
6
The study applies interpretable machine learning to make hedonic real-estate models more useful for inference and out-of-sample analysis.
Research Object
real estate rental markets and hedonic characteristics of apartments and buildings
Research Subject
the nonlinear effects, interactions, spatial relationships, and temporal variation of hedonic characteristics on rents and rental-market performance
Publication Details
Publication Date
2022-05-31
Journal
Publisher
ISSN
Cited by
129
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest
References available in scid.ai10
R: A Language and Environment for Statistical Computing2000
XGBoost2016
Greedy function approximation: A gradient boosting machine.2001
The Elements of Statistical Learning: Data Mining, Inference, and Prediction2010
Towards A Rigorous Science of Interpretable Machine Learning2017
Peeking Inside the Black-Box: A Survey on Explainable Artificial Intelligence (XAI)2018
Machine Learning Interpretability: A Survey on Methods and Metrics2019
Hedonic Pricing Models: A Selective and Applied Review2002
Prediction accuracy in mass appraisal: a comparison of modern approaches2013
Estimation and updating methods for hedonic valuation2019
Cited by3
Model agnostic interpretable machine learning for residential property valuation2023
Comparing hedonic and machine learning models in residential property valuation: evidence from sub-market analysis in Italy2026
Performance Evaluation of Machine Learning Algorithms and the Hedonic Model in Property Valuation: Emphasizing Structural and Spatial Attributes2026