The RATS Software Forum

The vast majority of empirical work done with VAR's uses flat priors on the coefficients and Jeffrey's prior for the covariance matrix.

Dear Tom,

I hope this finds you well,

When I run the procedure @FEVDTABLE I got the following message:

"## M4. A memory request for an additional 32768 bytes cannot be satisfied"

Is that a problem of my PC, I run it for only one variable and for only two periods;

@MCFEVDTable(model=SNVAR_EG1,INCLUDE=||3||,horizons=||1,10||)

another question, What is the main gain of estimating a full VAR model using the SURGibbsSetup procedure? Is this related to the prior type or the structure of the Variance Covariance matrix?

Last question: I try to estimate the same model with Minnesota prior using the SURGibbsSetup procedure, I don't know if there is a code for this, I tried using the procedures @BVARBuildPriorMN and @BVARFinishPrior, but I don't know how to manage these procedures with SURGibbsSetup, but I'm not sure of what I'm doing.

Last thing: I can't find the procedure @BVARFinishPrior in the material sent.

Thank you in advance

Dear Tom,

I'm really sorry to nag, but do you have some guidance for my questions.

Thank you

Dear Tom,

I hope this finds you well,

I tried running the procedure @FEVDTABLE on more powerful PCs, but I continue getting the following message:

"## M4. A memory request for an additional 32768 bytes cannot be satisfied"

I run it for only one variable and for only two periods;

@MCFEVDTable(model=SNVAR_EG1,INCLUDE=||3||,horizons=||1,10||)

Would you kindly guide me.

Thank you in advance