State Dependence and Unobserved Heterogeneity in the Extensive Margin of Trade
We study the role and drivers of persistence in the extensive margin of bilateral trade. Motivated by a stylized heterogeneous firms model of international trade with market entry costs, we consider dynamic three-way fixed effects binary choice models and study the corresponding incidental parameter problem. The standard maximum likelihood estimator is consistent under asymptotics where all panel dimensions grow at a constant rate, but it has an asymptotic bias in its limiting distribution, invalidating inference even in situations where the bias appears to be small. Thus, we propose two different bias-corrected estimators. Monte Carlo simulations confirm their desirable statistical properties. We apply these estimators in a reassessment of the most commonly studied determinants of the extensive margin of trade. Both true state dependence and unobserved heterogeneity contribute considerably to trade persistence and taking this persistence into account matters significantly in identifying the effects of trade policies on the extensive margin.
Code (1)
Similar Papers 제목 키워드 기반
Identification of Auction Models Using Order Statistics
Auction data often contain information on only the most competitive bids as opposed to all bids. The usual measurement error approaches to unobserved heterogeneity are inapplicable due to dependence among order statistic…
Order Statistics Approaches to Unobserved Heterogeneity in Auctions
We establish nonparametric identification of auction models with continuous and nonseparable unobserved heterogeneity using three consecutive order statistics of bids. We then propose sieve maximum likelihood estimators …
Identification of Average Marginal Effects in Fixed Effects Dynamic Discrete Choice Models
In nonlinear panel data models, fixed effects methods are often criticized because they cannot identify average marginal effects (AMEs) in short panels. The common argument is that identifying AMEs requires knowledge of …
Discrete Choice ModelsSpecification testing with grouped fixed effects
We propose a bootstrap generalized Hausman test for the correct specification of unobserved heterogeneity in both linear and nonlinear fixed-effects panel data models. We consider as null hypotheses two scenarios in whic…
Inference in High-Dimensional Panel Models: Two-Way Dependence and Unobserved Heterogeneity
Panel data allows for the modeling of unobserved heterogeneity, significantly raising the number of nuisance parameters and making high dimensionality a practical issue. Meanwhile, temporal and cross-sectional dependence…