The RATS Software Forum

Dear Tom,
Is there a way for imposing and testing a joint restriction on gamma and alpha coefficients in CATS?

CATS just does point estimates. One problem with doing Monte Carlo integration with cointegrated models is that the "beta" has a non-standard conditional distribution. If you're willing to take beta as given the remainder of the parameters can be handled using standard techniques---see th...

There are a lot of tweaks in the Fabiani-Mestre model (and most similar models) which are specific to the particular data set. (For instance, the PEGS is to override a problem with one of the variances going zero). If you post the data set, I can take a closer look. Thanks Tom. Here's my data set. ...

Hi, Is it possible to use the EXPORT option of @CATS in order to obtain the impulse-response graphs by the below procedures? @mcvardodraws(model=CatsModel) @mcgraphirf(model=CatsModel,page=one,varlabels=vlabels,shocklabels=vlabels, center=median,percent=||.025,.975||) Or is there any other way to ob...

Dear Tom, I tried to run ErrorBands.rpf of Fabiani & Mestre(2004) with annual series. However, I got the following error warning: The Error Occurred At Location 286, Line 12 of loop/block ## DLM1. Rank of Prediction Error Variance < Number of Observables The Error Occurred At Location 717, Line ...

Hi Tom,
Would you please explain what "-12.0*3" means in

compute sigmav(2)=%diag(%xdiag(sigmav(1)).*||.25,.25,4.0||)
compute logdetlimit=%MSSysRegInitVariances()-12.0*3

in the eev_ml.rpf? In what conditions should we change these two figures?
Thanks

TomDoan wrote:Not everything that's published is a good idea. Individual coefficients in a multi-lag dynamic model really don't have useful interpretations.

Certainly, I will take your word for it, Tom. Do you think that testing a Granger non-causality-like hypothesis is also useless?

Those are Wald tests on the results of the full switching model. You can use the Regression Tests wizard to set them up. I'm not sure what testing a difference between individual coefficients is going to tell you in a model like that. I just thought to make a table of results showing only the signi...

You can't test anything with the MCMC estimates---any testing has to be done with the maximum likelihood. You can do a LR test for covariance and coefficients switching vs just covariances by estimating the models with and without the coefficient switching. That will have standard asymptotics. Is t...

Yes. Though I would caution you (as I tell anyone who tries to adapt this), that the EEV program is rather specifically aimed at splitting the sample based upon the variance of the oil price. compute sigmav(2)=%diag(%xdiag(sigmav(1)).*||.25,.25,4.0||) is for the EEV data, as it guesses a 2nd regime...

You can't test anything with the MCMC estimates---any testing has to be done with the maximum likelihood. You can do a LR test for covariance and coefficients switching vs just covariances by estimating the models with and without the coefficient switching. That will have standard asymptotics. Is t...

You can't test anything with the MCMC estimates---any testing has to be done with the maximum likelihood. You can do a LR test for covariance and coefficients switching vs just covariances by estimating the models with and without the coefficient switching. That will have standard asymptotics. Then...

Dear Tom,
How can I test restrictions on BETASYS(k)(i,j) within the scope of eev_mcmc.rpf? For instance, I want to test if BETASYS(1)(i,j)= BETASYS(2)(i,j) or not. Can you please help me?
Many thanks.
Atarca

Thanks Tom. Then, is it a problem to have different "Draw -XXXX Executing swap" in each case?

Thank you very much for your quick reply, Tom. However, I wonder why it is %ran(.1) instead of being %ran(1.0) in this case.