Models - dummies and trends

For econometric discussions not necessarily related to EViews.

Moderators: EViews Gareth, EViews Moderator

ak1234
Posts: 7
Joined: Wed Feb 13, 2013 10:21 am

Models - dummies and trends

Postby ak1234 » Fri Mar 01, 2013 12:21 pm

Hi,

When making a model better can I use both dummy and trend in one model? - as this would then correct for structural break within my data also the trend will allow my model to then have the correct functional form.

I have already created separate models one including a dummy and the other a trend.

But I have to choose only two best long run models and the dummy model on its own does not give me the correct functional form.

Kind regards

A.

startz
Non-normality and collinearity are NOT problems!
Posts: 3798
Joined: Wed Sep 17, 2008 2:25 pm

Re: Models - dummies and trends

Postby startz » Fri Mar 01, 2013 1:05 pm

Yes.

ak1234
Posts: 7
Joined: Wed Feb 13, 2013 10:21 am

Re: Models - dummies and trends

Postby ak1234 » Fri Mar 01, 2013 1:11 pm

thank you.

A.

asiajade747
Posts: 1
Joined: Sun Mar 24, 2013 1:13 pm

Re: Models - dummies and trends

Postby asiajade747 » Sun Mar 24, 2013 1:17 pm

Hi Guys,
Hoping someone can help me...can you model a dummy variable to be case sensitive or does it have to be done manually? I want to create a dummy to correspond with my data sets increases and decreases.For instance, if the data increases it will be 0 and it will only be 1 if there are decreases year on year. How can I word this command?

startz
Non-normality and collinearity are NOT problems!
Posts: 3798
Joined: Wed Sep 17, 2008 2:25 pm

Re: Models - dummies and trends

Postby startz » Sun Mar 24, 2013 1:32 pm

Code: Select all

series dummy = @recode(x>x(-1),0,1)


Return to “Econometric Discussions”

Who is online

Users browsing this forum: No registered users and 2 guests