Machine Learning, Architectural Styles and Property Values

Машинное обучение, архитектурные стили и стоимость недвижимости
Thies Lindenthal, Erik B. Johnson
2021-07-13

architectural style classificationhedonic modelmachine learning classifiersmass appraisalproperty value premia
Abstract This paper couples a traditional hedonic model with architectural style classifications from human experts and machine learning (ML) enabled classifiers to estimate sales price premia over architectural styles, both at the building and the neighborhood-level. We find statistically and economically significant price differences for houses from distinct architectural styles across an array of specifications and modeling assumptions. Comparisons between classifications from ML models and human experts illustrate the conditions under which ML classifiers may perform at least as reliable as human experts in mass appraisal models. Hedonic estimates illustrate that the impact of architectural style on price is attenuated by properties with less well-defined styles and we find no evidence for differential price effects of Revival or Contemporary architecture for new construction.
1
Architectural-style effects on prices are weaker for properties with less clearly defined styles.
2
For new construction, the study finds no evidence of distinct price effects for Revival or Contemporary architecture.
3
Homes representing different architectural styles exhibit statistically and economically significant sales-price differences across multiple model specifications and assumptions.
4
Machine-learning architectural-style classifiers can perform at least as reliably as human experts under certain conditions in mass-appraisal models.
5
The study combines hedonic pricing models with human and machine-learning architectural-style classifications to estimate building- and neighborhood-level price premia.

Residential properties and neighborhoods classified by architectural style

The effects of architectural style on sales prices and the reliability of machine-learning versus human style classifications in hedonic mass-appraisal models

Publication Details
Publication Date
2021-07-13
Journal
Publisher
ISSN
Cited by
49
Access Type
Author Information
Authors
Thies Lindenthal
Erik B. Johnson
Explore further
Open the scid.ai AI chat with a ready-made request: it will find papers on a similar topic and help build a literature review.
Find similar papers in the chat →
Make a presentation
100%