Comparing automated valuation models for real estate assessment in the Santiago Metropolitan Region: A study on machine learning algorithms and hedonic pricing with spatial adjustments
Сравнение автоматизированных моделей оценки недвижимости в Столичном регионе Сантьяго: исследование алгоритмов машинного обучения и гедонического ценообразования с пространственными корректировками
2025-03-25
SCID: 54.1/a6kduajm
Discuss with AI
LightGBMSHAPautomated valuation modelsconvolutional neural networksspatial autoregressive model
Figures from the paper
Abstract (AI)
This study compares the precision and interpretability of two automated valuation models for evaluating the real estate market in the Santiago Metropolitan Region of Chile: machine learning algorithms, specifically LightGBM, and hedonic prices with spatial adjustments (SAR). Traditional residence attributes, such as housing amenities and proximity to services, were considered alongside visual information extracted from images using Convolutional Neural Networks (CNN). The research evaluates the influence of each model characteristic on performance metrics and identifies the relative importance of attributes using the SHapley Additive exPlanations (SHAP) algorithm. The results demonstrate the positive impact of image-based variables on performance metrics, showing that the introduction of visual information can considerably reduce error margins when estimating housing prices. In addition, the SHAP algorithm reveals complex non-linear interactions between price and crucial variables such as total surface area and neighborhood attributes, highlighting the importance of using methods that can capture these effects. Likewise, both LightGBM and SAR models indicate that variables that have the most significant impact on the value of properties are total surface area, municipality quality index, average academic level of nearby schools, and the number of bathrooms.
Key Findings
1
Both LightGBM and SAR identify total surface area, municipality quality, nearby schools’ average academic level, and bathroom count as the most influential valuation variables.
2
Combining traditional housing attributes with CNN-extracted visual information considerably reduces housing-price estimation errors.
3
SHAP analysis identifies complex nonlinear interactions involving total surface area, neighborhood characteristics, and property prices.
4
The findings support valuation methods capable of incorporating visual data and modeling spatial and nonlinear effects.
5
The study compares LightGBM machine-learning models with spatially adjusted hedonic pricing models (SAR) for property valuation in Santiago, Chile.
Research Object
Residential properties and the real estate market in the Santiago Metropolitan Region of Chile
Research Subject
The precision, interpretability, variable importance, and nonlinear effects of automated property valuation models using traditional, spatial, and image-based attributes
Publication Details
Publication Date
2025-03-25
Journal
Publisher
ISSN
Cited by
15
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest