# Residual standard error: 2.77 on 16710 degrees of freedom In order to obtain robust standard errors (i.e. in a similar way to vce(robust) in Stata) you can include robust = T to the arguments: # Residual deviance: 61989121 on 17083 degrees of freedom
# Null deviance: 78081855 on 17087 degrees of freedom # (Dispersion parameter for quasipoisson family taken to be 42366.31)
The resulting equation can be estimated by simple OLS.Īn example of how to apply the functions bvu and bvw to an example dataset in gravity and the resulting output is shown in the following: The transformed variables are included as independent variables in the estimation. When using weighted averages (BVW), the simple averages are replaced by GDP weights. X_\) the transformed variable adjusted for multilateral resistances. The underlying idea of a traditional gravity model, shown for international trade, is equally simple: \(G\) is a constant and as such of no major concern. The force \(F\) between two bodies \(i\) and \(j\) with \(i \neq j\) is proportional to the masses \(M\) of these bodies and inversely proportional to the square of their geographical distance \(D\). Gravity models in their traditional form are inspired by Newton law of gravitation:
0 kommentar(er)
