Evaluating Factor Contributions for Sold Homes

2025-11-03

AI Summary

This study examines the independent predictive contributions of ten internal and external factors—including environmental, social, and governance (ESG) metrics—to residential property prices. Using transaction data from three major U.S. cities (2022–2024), we develop a generalized additive model (GAM) enhanced with P-splines. We innovatively quantify marginal predictive power via adjusted R² change and correlation-controlled variable importance analysis. Results show that living area and geographic location exert the strongest influence; several ESG indicators—including energy efficiency ratings and age-friendly community infrastructure—exhibit statistically significant (p < 0.005) and robust incremental explanatory power, particularly in retirement-oriented cities where age-friendliness is especially salient. The proposed model outperforms standard benchmarks, and all predictors pass stringent significance tests. This work provides an interpretable, reproducible methodological framework and empirical evidence for integrating sustainability metrics into real estate valuation.

Technology Category

• Machine Learning: Efficient ML / Green AI
• Multiagent Systems: Mechanism Design
• Game Theory and Economic Paradigms: Mechanism Design

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Economics, Online Markets and Human Computation: Sustainability of Web economics
• User Modeling, Personalization and Recommendation: Metrics for user behavior and evaluating success

Abstract

We evaluate the contributions of ten intrinsic and extrinsic factors, including ESG (environmental, social, and governance) factors readily available from website data to individual home sale prices using a P-spline generalized additive model (GAM). We identify the relative significance of each factor by evaluating the change in adjusted R^2 value resulting from its removal from the model. We combine this with information from correlation matrices to identify the added predictive value of a factor. Based on data from 2022 through 2024 for three major U.S. cities, the GAM consistently achieved higher adjusted R^2 values across all cities (compared to a benchmark generalized linear model) and identified all factors as statistically significant at the 0.5% level. The tests revealed that living area and location (latitude, longitude) were the most significant factors; each independently adds predictive value. The ESG-related factors exhibited limited significance; two of them each adding independent predictive value. The elderly/disabled accessibility factor was much more significant in one retirement-oriented city. In all cities, the accessibility factor showed moderate correlation with one intrinsic factor. Despite the granularity of the ESG data, this study also represents a pivotal step toward integrating sustainability-related factors into predictive models for real estate valuation.

Problem

• Evaluating factor contributions to home sale prices using P-spline GAM
• Identifying relative significance of ESG factors in real estate valuation
• Assessing predictive value of intrinsic and extrinsic housing factors

Innovation

• P-spline generalized additive model evaluates housing factors
• Removing factors measures significance via adjusted R-squared
• Combining correlation matrices identifies independent predictive value