EViews.com

The most likely cause is that you do, in fact, have missing data over the forecast period. Best thing to do is post your workfile so we can take a look.

This actually can't be done in EViews 6 (at least I can't think of a way). You'll have to use EViews 7, where you can use the @name data member to retrieve the name.

The section "How to Create and Specify a System" of the "System Estimation" chapter of User Guide II gives pretty clear instructions on how to add AR terms to a system.

If you contact our office at sales@eviews.com they can help you with those questions.

With regards to the correlogram question, the easiest way to do it is to freeze the correlogram into a table, and then grab the p-value from that table. Something like: freeze(mytable) x.correl(6) !pval = mytable(7,7) As a general response to your post, you're probably best off reading through some ...

It really isn't clear what you are trying to do, or what you are saying. That code simply estimates 10 equations, and then displays the R2 from each. Nowhere does it create or open multiple workfiles.

The P-value should not be over 1. Either something has gone wrong with the "extracts", or something was wrong with EViews when the original estimations were done. I know that in the early days of EViews 6 there was a bug that reported the Durbin-Watson statistic (I think) as the p-value fo...

I think there are a few bugs in the implementation of this code when entering periods, rather than fractions of pi. The code: if (s > 1) then 'Allow the user to use periods when calling the routine. s1 = 2/e e = 2/s s = 2/e endif is definitely wrong, since s1 has not been declared anywhere. It isn't...

Well you can get indices by entering the symbol for those indices. The problem, though, is that most indices' symbols start with a "^" character (^GSPC for S&P500, I think), which is an illegal EViews name character. Thus the @recode part of my wfopen and import statements won't work. ...

Nothing has been added to 7 with regards to panel data estimation. Your specification looks like you're need to estimate fixed effects models, and the system object in EViews does not have options for fixed effects, although you could, possibly, always do them yourself with dummy variables.

Purely a personal recommendation, and in no way an official endorsement, but I like Compare it!

I think you need to be more clear with your definitions. EViews does not support the A-B Dynamic Panel System Estimator. That is completely separate from estimating a system of equations in a panel workfile, which is what your post appears to be about. Systems of equations can be estimated in a pane...

I think we'd have to see it in order to work out what is going wrong.

Could you send us your program and data?

Hard to say without seeing the data/specification.

I don't believe that is possible.