question about switching on @msvarsetup

Discussion of models with structural breaks or endogenous switching.

question about switching on @msvarsetup

Postby hardmann » Sun Jan 29, 2012 5:02 am

Dear Tom:

Dear Tom, I am sorry for I have some stupid questions.
When I run hamilton.rpf, I change option switch=mh or switch=ih on @msvarsetup. Becouse H indicates variances are state-dependent, why I always get one sigma other than two.


Thanks.
hardmann
 
Posts: 48
Joined: Sat Feb 26, 2011 10:49 pm

Re: question about switching on @msvarsetup

Postby TomDoan » Mon Jan 30, 2012 11:27 am

You have to use SIGMAV and not SIGMA in your NONLIN instruction if the variances are switching.
TomDoan
 
Posts: 2720
Joined: Wed Nov 01, 2006 5:36 pm


Return to Structural Breaks and Switching Models

Who is online

Users browsing this forum: No registered users and 1 guest