Page 4 of 4
Posted: Tue Oct 10, 2017 7:43 am
Is it possible to not include an intercept term? Some papers I've read that use daily data, suggest an intercept in both the linear and non-linear parts of this model are not needed because of the frequency.
Posted: Wed Oct 11, 2017 11:10 am
I have design the add-in in such a way that the intercept must be included in the model. But you can easily do that in the STAR EViews 10 procedure.
Posted: Mon Dec 11, 2017 10:46 am
I am writing my master thesis in financial economics and I am using the logistic smooth transition regression model in a factor model. I downloaded your add-in. When i use it via the window of my dependent variable or I use the command that I can find in your pdf, I only obtained scalars (with one value of 1 or 0 ) for the estimation and evaluation parts. I was wondering if you could help me with that.
These are the commands I tried to use:
exc1.star(variables= unffr mktrf ,transition=mpu,lstr,evaluation,sv=sv)
exc1.star(variables= unffr mktrf mpu ,transition=mpu,lstr,evaluation,sv=sv)
Posted: Thu Dec 21, 2017 6:08 pm
I guess that the estimation of your gamma parameter is large enough, hence you can use a discrete TAR model and it is not necessary to perform a LSTAR estimation.
Posted: Thu Jul 05, 2018 2:32 am
For the example 1, after installing the add-in just entee in the command window:
Code: Select all
z.star(variables=x x(-1) x(-2),transition=x(-1),lstr,evaluation,sv=sv)
¿Which version of Eviews are you using? ¿Of which fields to fill do you have questions? all of them are explained if the pdf.
Dear Nicolas, thanks for the STAR* add ins! I am having a problem with adding lags. What is right code of adding 10 lags? I used x(-1 to -10) but I get a mistake massage. Best regards.
Posted: Thu Jul 05, 2018 5:28 pm
You can not add the lags using that command. You must type each lag individually.