Dueker (1997) output with fixed mean

Discussion of models with structural breaks or endogenous switching.
TomDoan
Posts: 7814
Joined: Wed Nov 01, 2006 4:36 pm

Re: Dueker (1997) output with fixed mean

Unread post by TomDoan »

The C, A and B are from the preliminary fixed GARCH that also is used only for guess values. However, they correspond to gamma, alpha and beta.

In the main MAXIMIZE output, MSG(1) and MSG(2) are his alpha and beta. GV(2) is his gamma and GV(1) is his gamma x g(St=0). He parameterizes the degrees of freedom as 1/standard degrees of freedom, while we use degrees of freedom directly.
prashantj
Posts: 88
Joined: Sun Apr 11, 2010 2:56 am

Re: Dueker (1997) output with fixed mean

Unread post by prashantj »

TomDoan wrote:The C, A and B are from the preliminary fixed GARCH that also is used only for guess values. However, they correspond to gamma, alpha and beta.

In the main MAXIMIZE output, MSG(1) and MSG(2) are his alpha and beta. GV(2) is his gamma and GV(1) is his gamma x g(St=0). He parameterizes the degrees of freedom as 1/standard degrees of freedom, while we use degrees of freedom directly.
thanks..what do MU(1), MU(2), NUV(1), NUV(2), P(1,1) and P(1,2) suggest? It would be great if you recommend some materials which discuss in detail the interpretation of each output parameter. please
Looking forward for your reply,
Regards,
Prashant
TomDoan
Posts: 7814
Joined: Wed Nov 01, 2006 4:36 pm

Re: Dueker (1997) output with fixed mean

Unread post by TomDoan »

1. The paper.
2. The RATS manual and Switching course materials
3. http://en.wikipedia.org/wiki/Greek_alphabet

I don't mean to be flip, but I have no idea how else to respond to that.
Post Reply