Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/97149 
Title (translated): 
Gravity Modell Schätzung: Fixed Effects vs.Random Intercept Poisson Pseudo Maximum Likelihood
Year of Publication: 
2014
Series/Report no.: 
Discussion Paper No. 148
Publisher: 
Leibniz Institute of Agricultural Development in Transition Economies (IAMO), Halle (Saale)
Abstract: 
Since the work of FEENSTRA (2002), the standard ANDERSON & VAN WINCOOP (2003) Gravity Model has been estimated using a fixed effects approach. However, a fixed effects approach has a major drawback: it does not allow for the estimation of exporter- and importer-invariant variables. Thus, economically relevant variables such as exporter and importer gross domestic product are disregarded. Here, we propose a random intercept model to address this gap. This approach not only provides identical estimates to a fixed effects approach, but also allows for the estimation of exporter- and importer-invariant variables.
Abstract (Translated): 
Seit FEENSTRA (2002) wird das Standard ANDERSON & VAN WINCOOP (2003) Gravity Modell mittels Fixed-Effects-Ansatzes geschätzt. Der Fixed-Effects-Ansatz hat allerdings den entscheidenden Nachteil, dass er nicht die Schätzung von Exporteur- und Importeur-invarianten Variablen erlaubt. Folglich lassen sich ökonomisch relevante Variablen wie Exporteur- und Importeur-Bruttosozialprodukt nicht berücksichtigen. Wir empfehlen ein Random-Intercept-Modell anstatt. Dieser Ansatz liefert nicht nur die identischen Schätzer wie ein Fixed-Effects-Modell, sondern erlaubt auch die Schätzung von Exporteur- und Importeur-invarianten Variablen.
Subjects: 
Gravity Model Estimation
Poisson Pseudo Maximum Likelihood
Fixed Effects Model
Random Intercept Model
JEL: 
F1
C3
Persistent Identifier of the first edition: 
Document Type: 
Working Paper

Files in This Item:
File
Size





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.