Testing for Spatial Effects in Seemingly Unrelated Regressions

摘要: Abstract The paper focuses on the case of a panel data set, without unobserved individual effects, treated by means an SUR specification. problem raised is to test for presence spatial effects in these multivariate systems. Various useful tests are developed based principle Lagrange Multiplier maximum-likelihood framework. Also, we address question time stability sequence dependence coefficients, as maintained hypothesis that not necessarily true applied work. second part presents results Monte Carlo experiment. Essais sur les effets spatiaux dans des regressions apparemment sans rapport Resume Cette communication se concentre le cas de l'ensemble donnees panel, individuels non observes, traitees au moyen d'une specification SUR. Le probleme souleve concerne l'examen la d'effets ces systemes multi-variables. On developpe plusieurs essais utiles, bases pri...