Institution profile

Institut d'Etudes Politiques de Paris (Sciences Po)

Academic institutioneurope · fr
Official website
Research library18linked papers
Opportunities0open roles
Selected work

Representative Papers

Ridge Estimation of High Dimensional Two-Way Fixed Effect Regression

Jan 07, 2026

This study addresses the challenge of controlling bias and variance in estimating high-dimensional two-way fixed effects regression models under sparse bipartite networks. To this end, the authors propose a ridge regression–based regularization approach that stabilizes the estimation of fixed effect vectors by setting the regularization parameter to grow logarithmically with network size. Theoretical analysis demonstrates that both the bias and the covariance matrix of the proposed estimator converge to deterministic equivalents determined solely by the expected network structure. By integrating concentration inequalities, high-dimensional statistical inference, and sparse network modeling techniques, the work establishes the asymptotic properties of the estimator and validates its effectiveness and robustness through extensive simulations and empirical analysis using real administrative wage data.

1 citationsRead paper
Recent publications

Latest Papers

Using Pre-Trends for Inference in Difference-in-Differences

Jul 23, 2026

This study addresses the vulnerability of conventional difference-in-differences (DID) estimators to bias under post-treatment shocks, which arises from their reliance on the parallel trends assumption. To overcome this limitation, the authors propose a novel inference approach that dispenses with this assumption by constructing a DID-specific predictor based on pre-treatment outcome dynamics and embedding it within a conformal inference framework. This method explicitly models potential post-treatment shocks and leverages pre-treatment information to impose identification constraints, thereby enabling robust causal inference even when parallel trends fail to hold. The proposed procedure substantially enhances the reliability and applicability of DID estimates in settings characterized by non-parallel trends.

0 citationsRead paper