estimation: monthly model vs quarterly data

[jduras]

Hi,

I was wondering, is there is a way in Dynare to estimate parameters of a monthly model if the data that I try to match are quarterly time series?
Thanks!

Jan

[SébastienVillemot]

Yes, here is a broad outline of the solution (I have not tested it but I think it works):

• create a monthly dataset to be used for the Dynare estimation command
• if you have some observable vars on a monthly basis, enter them in the dataset in the normal way
• for the observables which you have on a quarterly basis, enter them only for the months of March, June, September and December, and leave NaN in the other months
• in your model, create new variables and equations for the quarterly variables (for example: GDP_QUARTERLY=GDP_MONTHLY+GDP_MONTHLY(-1)+GDP_MONTHLY(-2);)
• when you run the estimation, Dynare is able to nicely handle the missing values in the dataset

Hope this helps,