Dynamic or static panel data

[Joost Huijbregts]

Dear members,

I am new to this forum and currently working on my masters thesis.
For this, I want to examine the impact of certain political-economic variables on nominal exchange rates.
I am performing a panel data study and I have collected data from over 150 countries between 1980-2012.

However, I am struggling with my exchange rate variable. I use the difference in natural logarithms between two years as my dependent variable.
For my independent variables I use for example CBI, Exchnage rate regime and some macro-economic variables.
However, I am told that I have to include the lagged nominal exchange rate as well (which is very logical of course). Moreover, I have to do this with a B-1 variable.

LN∆eit = α + β1X_it-1+(B2-1)LNe_it-1 + εt in which X_it-1 are a set of explanatory variables and e_it-1 the lagged nominal exchange rate.

I have two questions.
How do I make this variable B-1 variable and why is it like that?
Is this a dynamic panel data model? I know that a dynamic panel data model includes a lagged dependent variable, but I am not sure if a lagged exchange rate counts as a lagged dependent variable if my dependent variable is a difference.

Any comments are much appreciated!

[Phil Bromiley]

Change variables are a little tricky - the parameter on x is the change in predicted change for a one unit change in x. If you're going to include prior exchange rate anyway, then why not just use exchange rate? [To estimate such models, the program often differences everything anyway.]

First, you xtset your data so you can use lag (L.) operators. Then you'll just need to include L.exchangerate in your model. However, if you include a lagged exchange rate in a model with exchange rate as the dv (or change in exchange rate), you'll need to take care of endogeneity. This can be done with xtivreg, xtivreg2, the xtabond estimators, or user written xtdpdml.

[Lahcen OULHAJ]

xtdpdml performs very well when T = 10 for any rho in (0, +1). How to run xtdpdml for larger T?