Estimating network spillovers under dense measurement error

Yingxing Li, Ã ureo de Paula and Weining Wang

No 21/26, CeMMAP working papers from Institute for Fiscal Studies

Abstract: This paper analyzes spillover effects in spatial (network) models when the neighborhood (adjacency) matrix is contaminated by measurement error from reporting, imputation, or disclosure imperfections, leading to inconsistent estimation of network effects. We introduce a regularization framework for the latent network that allows for sparse and/or low-rank structure and accommodates potential correlation between measurement errors and outcomes. We propose two estimators: (i) a two-stage procedure that first denoises the adjacency matrix and then incorporates the denoised network into a regression analysis, and (ii) a Generalized Method of Moments (GMM) estimator that jointly estimates regression parameters and refines the network structure. We establish consistency, convergence rates and asymptotic normality for the proposed spillover-effect estimators and show that their error bounds are strictly smaller than those of naive estimation, which may be inconsistent when measurement error does not vanish. Simulations demonstrate that, in the presence of noisy networks, our approach could substantially reduce the estimation errors relative to conventional methods. We apply our framework to examine international spillovers in economic growth, and tax competition across U.S. states.

Date: 2026-11-03
References: Add references at CitEc
Citations:

Downloads: (external link)
https://cemmap.ac.uk/wp-content/uploads/2026/11/CW ... easurement-error.pdf (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:azt:cemmap:21/26

DOI: 10.47004/wp.cem.2026.2126

Access Statistics for this paper

More papers in CeMMAP working papers from Institute for Fiscal Studies Contact information at EDIRC.
Bibliographic data for series maintained by Dermot Watson ().