forecasting exogenous variables

For technical support, tips and tricks, suggestions, or any other information regarding the EViews model object.

Moderators: EViews Gareth, EViews Moderator

Irishka
Posts: 4
Joined: Tue Oct 15, 2013 3:30 pm

forecasting exogenous variables

Postby Irishka » Tue Oct 15, 2013 3:55 pm

Hello dear Eviews Helpers! Desperately need your help and will be greatful for any advice!
I have made a model using VEC in which I have exogenous variables, I can solve the model, but when it gets to forecasting I get confused, I can easily forecast endogenous variables, but don't understand how to treat exogenous in order to be able to solve the model for the forecasted period. Thanks in advance!

EViews Gareth
Fe ddaethom, fe welon, fe amcangyfrifon
Posts: 13305
Joined: Tue Sep 16, 2008 5:38 pm

Re: forecasting exogenous variables

Postby EViews Gareth » Tue Oct 15, 2013 4:15 pm

They're exogenous. They can't be forecasted by the model (that's what exogenous means!). So how would you like to treat them?
Follow us on Twitter @IHSEViews

Irishka
Posts: 4
Joined: Tue Oct 15, 2013 3:30 pm

Re: forecasting exogenous variables

Postby Irishka » Wed Oct 16, 2013 2:21 am

Thank you for a quick reply, I have probably explained my problem incorrectly. I have actual monthly data from 2002 to 2012, I make a model based on VEC results, solve my model to see if it works (deterministic-static solution), then check the same period for dynamic solution, change my sample to 2020 to get a forecast and I can't solve the model in which I have 3 exogenous variables, it says missing data... if I change all exogenous to endogenous then I get a forecast, but it looks weird. So my question is how to perform a forecast with exogenous variables, so there would be no failure due to missing data. (For forecast, I solve stochastic-dynamic solution).Thanks

startz
Non-normality and collinearity are NOT problems!
Posts: 3775
Joined: Wed Sep 17, 2008 2:25 pm

Re: forecasting exogenous variables

Postby startz » Wed Oct 16, 2013 6:30 am

Irishka wrote:Thank you for a quick reply, I have probably explained my problem incorrectly. I have actual monthly data from 2002 to 2012, I make a model based on VEC results, solve my model to see if it works (deterministic-static solution), then check the same period for dynamic solution, change my sample to 2020 to get a forecast and I can't solve the model in which I have 3 exogenous variables, it says missing data... if I change all exogenous to endogenous then I get a forecast, but it looks weird. So my question is how to perform a forecast with exogenous variables, so there would be no failure due to missing data. (For forecast, I solve stochastic-dynamic solution).Thanks

That's because the values of the exogenous variables from 2013 to 2020 are missing. You need to supply values for these variables.

Irishka
Posts: 4
Joined: Tue Oct 15, 2013 3:30 pm

Re: forecasting exogenous variables

Postby Irishka » Wed Oct 16, 2013 8:42 am

but I don't understand how to get the data for the exogenous variables from 2012 to 2020 if it's a forecast? Do I make them up? write somewhere? could someone please provide an instruction on how I can feel that missing data for my exogenous variables! Thanks a lot!

EViews Gareth
Fe ddaethom, fe welon, fe amcangyfrifon
Posts: 13305
Joined: Tue Sep 16, 2008 5:38 pm

Re: forecasting exogenous variables

Postby EViews Gareth » Wed Oct 16, 2013 8:51 am

There is no right answer. By stating that they are exogenous, you are stating that they are determined outside of the model. How they are determined is entirely up to you.
Follow us on Twitter @IHSEViews

Irishka
Posts: 4
Joined: Tue Oct 15, 2013 3:30 pm

Re: forecasting exogenous variables

Postby Irishka » Wed Oct 16, 2013 9:08 am

So, from what I understood from the replies: I need to fill the missing data for my exogenous variables in order to solve my model up until 2020. But how????I tried creating a model with all endogenous variables and then just exclude, so that they will be treated as exogenous and created a scenario, but it didn't work and I am really confused. Can you write a step-by step instruction on what to do to fill my data?please&

EViews Gareth
Fe ddaethom, fe welon, fe amcangyfrifon
Posts: 13305
Joined: Tue Sep 16, 2008 5:38 pm

Re: forecasting exogenous variables

Postby EViews Gareth » Wed Oct 16, 2013 9:13 am

Open up each of the exogenous series, click the edit button and enter the numbers you want one by one.
Follow us on Twitter @IHSEViews


Return to “Models”

Who is online

Users browsing this forum: No registered users and 14 guests