Paper: arXiv 2404.07132

Authors: Jason R. Bailey, W. Brent Lindquist, Svetlozar T. Rachev

Abstract

Using data from 2000 through 2022, we analyze the predictive capability of the annual numbers of new home constructions and four available environmental, social, and governance factors on the average annual price of homes sold in eight major U.S. cities. We contrast the predictive capability of a P-spline generalized additive model (GAM) against a strictly linear version of the commonly used generalized linear model (GLM). As the data for the annual price and predictor variables constitute non-stationary time series, to avoid spurious correlations in the analysis we transform each time series appropriately to produce stationary series for use in the GAM and GLM models. While arithmetic returns or first differences are adequate transformations for the predictor variables, for the average price response variable we utilize the series of innovations obtained from AR(q)-ARCH(1) fits. Based on the GAM results, we find that the influence of ESG factors varies markedly by city, reflecting geographic diversity. Notably, the presence of air conditioning emerges as a strong factor. Despite limitations on the length of available time series, this study represents a pivotal step toward integrating ESG considerations into predictive real estate models.

Complexity vs Empirical Score

  • Math Complexity: 7.5/10
  • Empirical Rigor: 4.0/10
  • Quadrant: Lab Rats — high math complexity, limited empirical testing

Why this score: The paper employs advanced statistical methods including P-spline Generalized Additive Models (GAMs), AR(q)-ARCH(1) fits for stationarity, and principal component analysis on residuals, indicating high mathematical density. However, empirical rigor is moderate due to a small sample size (23 years of annual data for 8 cities) and a lack of detailed backtesting or robustness checks typical of production-ready trading models.

Research Flowchart

  flowchart TD
  A["Research Goal<br>Predict home prices using ESG factors and construction data"] --> B{"Data Processing"}
  B --> C["Data Input: 2000-2022 US Home Prices & ESG Data"]
  C --> D["Stationarity Transformation<br>First Diffs/Returns & AR(q)-ARCH(1) innovations"]
  D --> E["Computational Modeling"]
  E --> F["GAM: P-spline Model"]
  E --> G["GLM: Linear Model"]
  F --> H{"Key Findings / Outcomes"}
  G --> H
  H --> I["ESG influence varies by city"]
  H --> J["Air conditioning is a strong factor"]
  H --> K["Time series transformation prevents spurious correlation"]

Evidence Card

What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card

Proposes

The paper proposes a predictive hedonic model using a P-spline GAM versus a linear GLM to assess ESG factors on average annual home prices.

Source sentence
Using data from 2000 through 2022, we analyze the predictive capability of the annual numbers of new home constructions and four available environmental, social, and governance factors on the average annual price of homes sold in eight major U.S. cities.

Assumes 3

  • Non‑stationary time series must be transformed to stationary form to avoid spurious correlations. statistical
    Source sentence
    As the data for the annual price and predictor variables constitute non-stationary time series, to avoid spurious correlations in the analysis we transform each time series appropriately to produce stationary series for use in the GAM and GLM models.
  • The model is deliberately parsimonious, implying latent variables may exist. model
    Source sentence
    As our model is deliberately parsimonious in terms of the number of predictor factors, we hypothesize the presence of such latent variables.
  • Cross‑sectional hedonic models assume i.i.d. samples, which is not applicable to the time‑series setting used here. data
    Source sentence
    Implicit in the cross-sectional analysis is the assumption that the data set represents independent, identically distributed, random samples reflective of the pricing structure in a particular geographic area.

Data

  • Zillow home sale data · Zillow market-prices public
    2000-2022 · Eight major U.S. cities (Atlanta, Austin, Columbus, Jacksonville, Nashville, Oklahoma City
    Source sentence
    Price and factor data were acquired from Zillow1. Our data set is composed of completed sale transactions of homes

Costs

Not applicable The paper reports no trading or execution result.

Finds 4

  • The GAM provides a model that is superior to the GLM in predictive performance. predictive-accuracy
    Source sentence
    the GAM provides a model that is superior to the GLM.
  • Central air conditioning is a strong ESG factor influencing home prices across cities. statistical
    Source sentence
    Notably, the presence of air conditioning emerges as a strong factor.
  • The influence of ESG factors varies markedly by city, reflecting geographic diversity. statistical
    Source sentence
    Based on the GAM results, we find that the influence of ESG factors varies markedly by city, reflecting geographic diversity.
  • Residual analysis indicates systemic factors not included in the models exist in the residuals. statistical
    Source sentence
    we conclude that systemic factors not included in our models exist in the residuals.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 38 related papers). Coverage is partial, so this is neither replication nor refutation.

Code

No repository by the authors found in the paper.

Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.

Cited References (10)

  1. When climate meets real estate: A survey of the literature (2024) — link
  2. Hedonic Models of Real Estate Prices: GAM Models; Environmental and Sex-Offender-Proximity Factors (2022) — link
  3. A Natural Disasters Index (2020) — link
  4. Environmental risk and differentiated housing values: Evidence from the north of France (2019) — link
  5. Research on the complexity of green innovative enterprise in dynamic game model and governmental policy making (2019) — link
  6. Flexible smoothing with B-splines and penalties (1996) — link
  7. Financial Econometrics (2022) — link
  8. Energy-relevant decisions of home buyers in new home construction (2013)
and 2 more
  1. 2010 U. S. Census (2009)
  2. Extreme value theory : an introduction (2006) — link

0 of 10 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.