On Wed, 8 May 2019, Lee Adkins wrote:
Did I miss something? I take it that the bkw package has been
"integrated"
into gretl making it unnecessary? That is fine with me. Is there something
I need to do?
I think you're right, Lee. The "vif" command now supports the new
$result accessor. As per the doc:
"Following this command the $result accessor may be used to retrieve a
bundle containing two matrices: a column vector holding the VIFs under
the key vif and the BKW matrix under BKW."
So I'm not sure there's anything left to do here.
Allin
> On Wed, May 8, 2019 at 1:14 AM Sven Schreiber <svetosch(a)gmx.net> wrote:
>
>> Am 08.05.2019 um 01:22 schrieb Allin Cottrell:
>>> So far as I'm concerned we're about ready for the gretl 2019b
release.
>>> Does anyone have any bugs they want to report -- or do any developers
>>> have new stuff they'd like to add -- before we do that?
>>
>> Spontaneously I remember the VIF stuff. There is a ticket on the tracker
>> about it:
https://sourceforge.net/p/gretl/feature-requests/101/
>> with a suggestion from me two months ago.
>>
>> I also contacted Lee about the situation in connection with his package
>> (and non-linear estimators I think), but have received no reply.
>>
>> Then there's the thread "Lee's bkw.gfn and vif (was Re: (now
>> vif-related) Re: new accessor)", also in March.
>>
>> It ended with you (Allin) saying "OK, we can try some experimentation."