The RATS Software Forum

Hi

I would like to use the GVAR model to analyse my data on trade in Africa. I know one can use any software Eviews, Stata and so on to run the regression. I have been advised to use SAS, Rats or Mathlab, but I don't know where to start, I've already collected all the data.

In the literature usually the data use is quarterly data,but I am using annual data since the quarterly one are not available in most African countries.

please advise,

Emilie

There's nothing particularly complicated about the technical aspects of the global VAR, it just requires a lot of careful data gathering. Have you figured out what weights you're going to use?

Hi tom

The paper we are following computed the weight using the annual trade flows average over a period (2006-2008 they used quarterly data). since I am using annual data I would like to compute the weight using a longer period let assume from 2000 to 2011.

I am studying linkage between sub region in Africa, Should I compute weight for each sub region. sorry if y questions look stupid it is my first time to use panel data.
regards

emi_kinf wrote:Hi tom

The paper we are following computed the weight using the annual trade flows average over a period (2006-2008 they used quarterly data). since I am using annual data I would like to compute the weight using a longer period let assume from 2000 to 2011.

I am studying linkage between sub region in Africa, Should I compute weight for each sub region. sorry if y questions look stupid it is my first time to use panel data.
regards

So your question is more about how you compute the weights from the data you have rather than how to do the global VAR given the weights. Is that correct? What data do you have?