2:44 p.m.
A follow-up to
http://lists.wfu.edu/pipermail/gretl-devel/2015-July/005882.html
(which still applies as of this writing).
Some new things in the gretl source since sourceforge CVS went down:
* Fix the bug noted by Sven in
http://lists.wfu.edu/pipermail/gretl-devel/2015-July/005880.html
* Add a new square() function (documented) with an optional argument
to generate cross-products as well as squares. Retire the old xpx()
function but preserve the name as an alias for square() with the
cross-product option activated.
* Add a seasonals() function (documented) with two optional
arguments, the first to specify a period to omit and the second
to specify centering of the seasonal dummies.
* "vif" command: lose the not-very-useful info on X'X that we've
printed to date, and substitute the Belsley-Kuh-Welsch
variance decomposition table, as advocated by Lee Adkins at
the Berlin gretl conference.
* GUI function package editor: add check for unsaved changes under
"Extra properties". Do more checking and give more feedback when
saving a package in zip format.
* Function package browser: add a toolbar item ("index" icon, tooltip
"Package registry") to show which packages are registered to
appear on which menus, with the option of removing them.
* Numerous "internal" tweaks to GUI function package apparatus in
the interest of greater robustness.
* Overwrite check on saving to file: switch to the built-in GTK API,
maybe catch more cases where overwriting should be flagged?
* Updated Romanian translation (thanks, Mihaela!)
Allin
4:52 p.m.
Allin,
This all looks great. However, when trying out the new 'vif' command, the variance
decomposition table is all 0's (both gui and console). This is on Windows 7.
<hansl>
open andy.gdt # in POE4
ols sales const price advert
vif
</hansl>
PS
-----Original Message-----
From: gretl-devel-bounces(a)lists.wfu.edu [mailto:gretl-devel-
bounces(a)lists.wfu.edu] On Behalf Of Allin Cottrell
Sent: Tuesday, July 21, 2015 2:45 PM
To: Gretl development
Subject: [Gretl-devel] source news
A follow-up to
http://lists.wfu.edu/pipermail/gretl-devel/2015-July/005882.html
(which still applies as of this writing).
Some new things in the gretl source since sourceforge CVS went down:
* Fix the bug noted by Sven in
http://lists.wfu.edu/pipermail/gretl-devel/2015-July/005880.html
* Add a new square() function (documented) with an optional argument
to generate cross-products as well as squares. Retire the old xpx()
function but preserve the name as an alias for square() with the
cross-product option activated.
* Add a seasonals() function (documented) with two optional
arguments, the first to specify a period to omit and the second
to specify centering of the seasonal dummies.
* "vif" command: lose the not-very-useful info on X'X that we've
printed to date, and substitute the Belsley-Kuh-Welsch
variance decomposition table, as advocated by Lee Adkins at
the Berlin gretl conference.
* GUI function package editor: add check for unsaved changes under
"Extra properties". Do more checking and give more feedback when
saving a package in zip format.
* Function package browser: add a toolbar item ("index" icon, tooltip
"Package registry") to show which packages are registered to
appear on which menus, with the option of removing them.
* Numerous "internal" tweaks to GUI function package apparatus in
the interest of greater robustness.
* Overwrite check on saving to file: switch to the built-in GTK API,
maybe catch more cases where overwriting should be flagged?
* Updated Romanian translation (thanks, Mihaela!)
Allin
_______________________________________________
Gretl-devel mailing list
Gretl-devel(a)lists.wfu.edu
http://lists.wfu.edu/mailman/listinfo/gretl-devel
5:08 p.m.
On Tue, 21 Jul 2015, Summers, Peter wrote:
Allin,
This all looks great. However, when trying out the new 'vif'
command, the variance decomposition table is all 0's (both gui and
console). This is on Windows 7.
<hansl>
open andy.gdt # in POE4
ols sales const price advert
vif
</hansl>
Hmm, not sure what's going wrong there. On Linux with current CVS
I'm seeing:
<gretl-output>
Belsley-Kuh-Welsch collinearity diagnostics:
--- variance proportions ---
lambda cond const price advert
2.880 1.000 0.001 0.001 0.019
0.116 4.973 0.010 0.012 0.976
0.004 26.660 0.989 0.987 0.006
</gretl-output>
I'll try testing on Windows tomorrow. This seems an unlikely thing
to work differently across platforms, but maybe the OpenBLAS DLL
I've built for Windows is somehow wonky? Although in that case I'd
expect many things to be seriously wrong, not just the BKW
diagnostics.
This is on 64-bit Windows, right?
Allin
6:11 p.m.
This is on 64-bit Windows, right?
Roger that.
PS
Sent from my iPad
On Jul 21, 2015, at 5:08 PM, Allin Cottrell
<cottrell@wfu.edu<mailto:cottrell@wfu.edu>> wrote:
On Tue, 21 Jul 2015, Summers, Peter wrote:
Allin,
This all looks great. However, when trying out the new 'vif' command, the variance
decomposition table is all 0's (both gui and console). This is on Windows 7.
<hansl>
open andy.gdt # in POE4
ols sales const price advert
vif
</hansl>
Hmm, not sure what's going wrong there. On Linux with current CVS I'm seeing:
<gretl-output>
Belsley-Kuh-Welsch collinearity diagnostics:
--- variance proportions ---
lambda cond const price advert
2.880 1.000 0.001 0.001 0.019
0.116 4.973 0.010 0.012 0.976
0.004 26.660 0.989 0.987 0.006
</gretl-output>
I'll try testing on Windows tomorrow. This seems an unlikely thing to work differently
across platforms, but maybe the OpenBLAS DLL I've built for Windows is somehow wonky?
Although in that case I'd expect many things to be seriously wrong, not just the BKW
diagnostics.
This is on 64-bit Windows, right?
Allin
_______________________________________________
Gretl-devel mailing list
Gretl-devel@lists.wfu.edu<mailto:Gretl-devel@lists.wfu.edu>
http://lists.wfu.edu/mailman/listinfo/gretl-devel
10:23 a.m.
On Tue, 21 Jul 2015, Summers, Peter wrote:
Allin,
This all looks great. However, when trying out the new 'vif'
command, the variance decomposition table is all 0's (both gui and
console). This is on Windows 7.
Ah, the actual computations were fine, it was a bug in the printout
code. The fix will be in a snapshot later today.
Allin
3370
days inactive
3371
days old
4 comments
2 participants
participants (2)
-
Allin Cottrell -
Summers, Peter