Gravity model problem

[estrella bernabeu]

Hi everybody,

I'm using the gravity model in my empirical research and i want some help.

My model as you know is about bilateral trade, expressed according to GDP per capita, superficy, export, import, distance, exchange rate.
My problem is with the exchange rate data, i'm opering on Tunisia and its relation with the ohter countries, from 1990 to 2008, i have many european countries in my data, so the major problem is how can i explain my exchange rate data before the euro it means before 1999, by knowing that there are countrie until now are out the EU as Switzerland.

The second thing, i should I calculate the exchange rate balanced with regard to the exports??

many thanks for your help :D

[estrella bernabeu]

Hi,
Nobody here can help me ?? :( i'm in big problem and i'm runnig out of time so if there is anyone who wants to be volunteer i'll appreciate.

[masely]

Use real exchange rate (rer) instead of nominal one.
select your target country's rer as a baseline (=100) and rearrange other currencies subject to the change.
once you get the data sorted on excel, you can run it on eviews.
the dependent variable will be a function of rer and other explanatory variables.