STAR output issues

[EViews Glenn]

On the original wonky test output... we've identified an issue with the scaling of some collinearity diagnostics. We are testing a fix and hope to have something soon.

Many apologies, and thanks for your patience.

[somordha]

Glenn, hopefully you'll let me know if there's progress on that. Steve

[EViews Glenn]

Very sorry about the delay. I wanted to make sure that we sorted everything out correctly.

We've put a fix into the code and testing looks good. It will appear in the next patch (probably early next week).

Thanks for your patience, and I greatly apologize for the inconvenience.

Let me know if you have any other issues or if you are under time pressure.

[somordha]

Thanks Glenn, when the patch is out I'll tackle and revert.

Thanks for your help.

[EViews Glenn]

It's up. Take a look.

[somordha]

Checked out that patch Glenn. Seems to solved some of the problems but I ran on another dataset from a published paper to double check and can't replicate results. Can I PM you the output?

[somordha]

Also, should the delay parameter be entered as a positive or negative number? I get all NAs for errors and t values when I enter as a negative.