The Microeconometrics Seminar: Every Tuesday at 12:15 pm.
Time: 12:15 pm – 1:15 pm
Date: 20th of March 2018 (NB: exceptionally on Wednesday)
Place: Room 3001.
Yannick GUYONVARCH (CREST) – “Asymptotic Results for Dyadic Data”, joint work with Laurent Davezies and Xavier D’Haultfoeuille.

Abstract: Trade flows between countries, relationships in a network, results of sports matches can all be seen as “interaction data”, where units from a same population interact. Such data exhibit particular dependence patterns, affecting the statistical properties of the estimators. We establish in this context simple and uniform law of large numbers and central limit theorems, implying the asymptotic normality of a large class of linear and nonlinear estimators. We also derive consistent variance estimators and show that they can be computed easily with statistical software. We also show the general consistency of a particular bootstrap scheme. Monte Carlo simulations suggest that the two methods work well even with few observations. Finally, we revisit Santos Silva and Tenreyro (2006) and show that accounting for dependence patterns in trade data has potentially large effects on standard errors.

Organizers:
Laurent Davezies (CREST), Benoit Schmutz (CREST), Arne Uhlendorff (CREST) & Lucas Girard (CREST)
Sponsors:
CREST
Lunch registration:
food provided, no registration

The Microeconometrics Seminar: Every Tuesday at 12:15 pm.
Time: 12:15 pm – 1:15 pm
Date: 20th of March 2018 (NB: exceptionally on Wednesday)
Place: Room 3001.
Yannick GUYONVARCH (CREST) – “Asymptotic Results for Dyadic Data”, joint work with Laurent Davezies and Xavier D’Haultfoeuille.

Abstract: Trade flows between countries, relationships in a network, results of sports matches can all be seen as “interaction data”, where units from a same population interact. Such data exhibit particular dependence patterns, affecting the statistical properties of the estimators. We establish in this context simple and uniform law of large numbers and central limit theorems, implying the asymptotic normality of a large class of linear and nonlinear estimators. We also derive consistent variance estimators and show that they can be computed easily with statistical software. We also show the general consistency of a particular bootstrap scheme. Monte Carlo simulations suggest that the two methods work well even with few observations. Finally, we revisit Santos Silva and Tenreyro (2006) and show that accounting for dependence patterns in trade data has potentially large effects on standard errors.

Organizers:
Laurent Davezies (CREST), Benoit Schmutz (CREST), Arne Uhlendorff (CREST) & Lucas Girard (CREST)
Sponsors:
CREST
Lunch registration:
food provided, no registration