Dynamic Spatial Treatment Effects as Continuous Functionals: Theory and Evidence from Healthcare Access

Fuente: arXiv
Saved in:
Bibliographic Details
Main Author: Kikuchi, Tatsuru
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917044456259584
author Kikuchi, Tatsuru
author_facet Kikuchi, Tatsuru
contents I develop a continuous functional framework for spatial treatment effects grounded in Navier-Stokes partial differential equations. Rather than discrete treatment parameters, the framework characterizes treatment intensity as continuous functions $τ(\mathbf{x}, t)$ over space-time, enabling rigorous analysis of boundary evolution, spatial gradients, and cumulative exposure. Empirical validation using 32,520 U.S. ZIP codes demonstrates exponential spatial decay for healthcare access ($κ= 0.002837$ per km, $R^2 = 0.0129$) with detectable boundaries at 37.1 km. The framework successfully diagnoses when scope conditions hold: positive decay parameters validate diffusion assumptions near hospitals, while negative parameters correctly signal urban confounding effects. Heterogeneity analysis reveals 2-13 $\times$ stronger distance effects for elderly populations and substantial education gradients. Model selection strongly favors logarithmic decay over exponential ($Δ\text{AIC} > 10,000$), representing a middle ground between exponential and power-law decay. Applications span environmental economics, banking, and healthcare policy. The continuous functional framework provides predictive capability ($d^*(t) = ξ^* \sqrt{t}$), parameter sensitivity ($\partial d^*/\partial ν$), and diagnostic tests unavailable in traditional difference-in-differences approaches.
format Preprint
id arxiv_https___arxiv_org_abs_2510_15324
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Dynamic Spatial Treatment Effects as Continuous Functionals: Theory and Evidence from Healthcare Access
Kikuchi, Tatsuru
Econometrics
Theoretical Economics
Applications
Methodology
I develop a continuous functional framework for spatial treatment effects grounded in Navier-Stokes partial differential equations. Rather than discrete treatment parameters, the framework characterizes treatment intensity as continuous functions $τ(\mathbf{x}, t)$ over space-time, enabling rigorous analysis of boundary evolution, spatial gradients, and cumulative exposure. Empirical validation using 32,520 U.S. ZIP codes demonstrates exponential spatial decay for healthcare access ($κ= 0.002837$ per km, $R^2 = 0.0129$) with detectable boundaries at 37.1 km. The framework successfully diagnoses when scope conditions hold: positive decay parameters validate diffusion assumptions near hospitals, while negative parameters correctly signal urban confounding effects. Heterogeneity analysis reveals 2-13 $\times$ stronger distance effects for elderly populations and substantial education gradients. Model selection strongly favors logarithmic decay over exponential ($Δ\text{AIC} > 10,000$), representing a middle ground between exponential and power-law decay. Applications span environmental economics, banking, and healthcare policy. The continuous functional framework provides predictive capability ($d^*(t) = ξ^* \sqrt{t}$), parameter sensitivity ($\partial d^*/\partial ν$), and diagnostic tests unavailable in traditional difference-in-differences approaches.
title Dynamic Spatial Treatment Effects as Continuous Functionals: Theory and Evidence from Healthcare Access
topic Econometrics
Theoretical Economics
Applications
Methodology
url https://arxiv.org/abs/2510.15324