Does a dynamic forecast from a VAR model with exogenous variables consider the values of the exogenous variables in its first forecast

[Chris Collins]

Hi

I have a VAR model with 4 endogeous variables, 2 exogenous variables and 1 lag.

The code is as follows,

global y2 fdwti fdcoal strg_dev_fd fd_ng
global exog weathershockhdd storage_sf

var $y2 , lags(1) exog($exog)

Where fdwti, fdcoal, strg_dev_fd, fd_ng, weathershockhdd, and storage_sf are all variable.


I am generating a dynamic forecast using this model.


I set up the forecast using this code.

estimates store m1_1L_f
forecast create m1_1L_model, replace
forecast estimates m1_1L_f

I then run a 4 week dynamic forecast using this code

forecast solve, suffix(_forecast) begin(date2[`i']) periods(4)

So my question is as follows, when I generate the forecasts, do these forecasts ignore the values of the exogenous variables at the time that the forecast is made. Its obvious that it does not consider these values in the forecasts after the first period, I'm just not sure about the first one.

I noticed that with the forecast function you can specify exogenous variables, however as can be seen my model already included them, hence the confusion.

Thank you for any help.