The paper introduces GME-GNN, which replaces group intercepts with aggregated balancing statistics and GNN layers to enable cross-group comparisons while achieving double robustness and asymptotic normality under network confounding.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
econ.EM 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Graph Neural Networks for Generalized Mundlak Estimator under Network Confounding
The paper introduces GME-GNN, which replaces group intercepts with aggregated balancing statistics and GNN layers to enable cross-group comparisons while achieving double robustness and asymptotic normality under network confounding.