Overview
Jean-Marie Dufour is an econometrician at CIRANO, CIREQ, and the Département de Sciences Économiques at Université de Montréal. He is known for foundational work on finite-sample inference in econometric models, including the theory of Monte Carlo and Maximized Monte Carlo tests, and for establishing the conditions under which standard asymptotic procedures fail in VAR models.
Key Contributions / Features
- Dufour (2006) — "Monte Carlo Tests with Nuisance Parameters: A General Approach to Finite-Sample Inference and Nonstandard Asymptotics in Econometrics" (Journal of Econometrics 133): general theory of MC and MMC tests; the key identity that the MC p-value is exactly uniform under the null when α(N+1) is an integer.
- Dufour and Jouini (2006) — finite-sample simulation-based inference in VAR models; demonstration that asymptotic and bootstrap Granger causality tests have catastrophic size; MMC test as provably exact solution.
- Dufour and Renault (1998) — "Short Run and Long Run Causality in Time Series: Theory" (Econometrica 66): multi-horizon Granger causality; shows that direct zero restrictions on VAR coefficients are not equivalent to Granger non-causality in higher-dimensional systems.
Related