10 a.m.
I realize that in the middle of August we're in down-time for many
people, so recent gretl snapshots probably haven't had much of a
workout.
All the same, let me ask: has anyone found problems lately that should
be addressed before we release gretl 1.10.2? I'd like to do so before
the semester starts here (i.e. by the end of next week).
Allin
10:23 a.m.
Hi Allin,
I've been having some intermittent problems with the "panel plot" command.
Sometimes if I right-click on on a variable, select panel plot, then "multiple plots
in grid," things work ok. But doing the same for other variables generates an error:
"C:\Program Files\gretl\wgnuplot.exe"
"C:\Users\psummers\AppData\Roaming\gretl\gpttmp.e11692": exit code 1
Things work ok if I plot the time series by group (the default). This is with 1.10.90cvs,
built 7-31, on Windows 7 64-bit.
I tried sending an example data set (both on- and off-list), but your daemon doesn't
like the attachment...
Best,
Peter
-----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, August 18, 2015 10:01 AM
To: Gretl development
Subject: [Gretl-devel] next release?
I realize that in the middle of August we're in down-time for many people, so
recent gretl snapshots probably haven't had much of a workout.
All the same, let me ask: has anyone found problems lately that should be
addressed before we release gretl 1.10.2? I'd like to do so before the semester
starts here (i.e. by the end of next week).
Allin
_______________________________________________
Gretl-devel mailing list
Gretl-devel(a)lists.wfu.edu
http://lists.wfu.edu/mailman/listinfo/gretl-devel
12:16 p.m.
On Tue, 18 Aug 2015, Summers, Peter wrote:
I've been having some intermittent problems with the "panel
plot"
command. Sometimes if I right-click on on a variable, select panel
plot, then "multiple plots in grid," things work ok. But doing the
same for other variables generates an error: "C:\Program
Files\gretl\wgnuplot.exe"
"C:\Users\psummers\AppData\Roaming\gretl\gpttmp.e11692": exit code 1
Things work ok if I plot the time series by group (the default).
This is with 1.10.90cvs, built 7-31, on Windows 7 64-bit.
I tried sending an example data set (both on- and off-list), but
your daemon doesn't like the attachment...
Thanks, Peter. I'll take a look at this. Can you give me an idea of
the dimensions of the panel, N and T, and maybe some notion of what
the data are like (quantitative or dummies, strictly positive or
not...)?
Allin
> -----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, August 18, 2015 10:01 AM
> To: Gretl development
> Subject: [Gretl-devel] next release?
>
> I realize that in the middle of August we're in down-time for many people, so
> recent gretl snapshots probably haven't had much of a workout.
>
> All the same, let me ask: has anyone found problems lately that should be
> addressed before we release gretl 1.10.2? I'd like to do so before the semester
> starts here (i.e. by the end of next week).
12:25 p.m.
Thanks, Peter. I'll take a look at this. Can you give me an idea
of the dimensions
of the panel, N and T, and maybe some notion of what the data are like
(quantitative or dummies, strictly positive or not...)?
N = 14, T = 139 (unbalanced). Both variables are quantitative and non-negative. Below are
simple summary stats for the problematic variable, by country (unit). There's one that
has no valid observations -- could that be the problem?
PS
Summary statistics for govass, by value of iso
iso = AUS (n = 61):
Mean 17132.
Minimum 2351.0
Maximum 48959.
Standard deviation 13476.
Missing obs. 78
iso = CAN (n = 59):
Mean 55.000
Minimum 2.5440
Maximum 290.49
Standard deviation 62.856
Missing obs. 80
iso = CHE (n = 61):
Mean 35831.
Minimum 4231.0
Maximum 1.2819e+005
Standard deviation 25838.
Missing obs. 78
iso = DEU (n = 58):
Mean 561.30
Minimum 5.0600
Maximum 1529.3
Standard deviation 550.07
Missing obs. 81
iso = DNK (n = 61):
Mean 79.289
Minimum 0.53900
Maximum 389.10
Standard deviation 108.05
Missing obs. 78
iso = ESP (n = 57):
Mean 1.0849e+007
Minimum 47200.
Maximum 3.7300e+007
Standard deviation 1.2319e+007
Missing obs. 82
iso = FRA (n = 61):
Mean 718.98
Minimum 12.940
Maximum 3410.8
Standard deviation 1022.4
Missing obs. 78
Dropping govass: sample range contains no valid observations
iso = GBR (n = 0):
iso = ITA (n = 35):
Mean 3.2676e+005
Minimum 27612.
Maximum 5.2809e+005
Standard deviation 1.7338e+005
Missing obs. 104
iso = JPN (n = 39):
Mean 203.69
Minimum 14.553
Maximum 385.12
Standard deviation 109.09
Missing obs. 100
iso = NLD (n = 52):
Mean 1.0669e+005
Minimum 3981.0
Maximum 3.2605e+005
Standard deviation 1.0164e+005
Missing obs. 87
iso = NOR (n = 59):
Mean 74559.
Minimum 2280.0
Maximum 2.5719e+005
Standard deviation 78311.
Missing obs. 80
iso = SWE (n = 61):
Mean 88817.
Minimum 1192.0
Maximum 5.1913e+005
Standard deviation 1.0296e+005
Missing obs. 78
iso = USA (n = 59):
Mean 420.11
Minimum 90.400
Maximum 2229.0
Standard deviation 353.52
Missing obs. 80
-----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, August 18, 2015 12:17 PM
To: Gretl development
Subject: Re: [Gretl-devel] next release?
On Tue, 18 Aug 2015, Summers, Peter wrote:
> I've been having some intermittent problems with the "panel plot"
> command. Sometimes if I right-click on on a variable, select panel
> plot, then "multiple plots in grid," things work ok. But doing the
> same for other variables generates an error: "C:\Program
> Files\gretl\wgnuplot.exe"
> "C:\Users\psummers\AppData\Roaming\gretl\gpttmp.e11692": exit code 1
>
> Things work ok if I plot the time series by group (the default).
> This is with 1.10.90cvs, built 7-31, on Windows 7 64-bit.
>
> I tried sending an example data set (both on- and off-list), but your
> daemon doesn't like the attachment...
Thanks, Peter. I'll take a look at this. Can you give me an idea of the dimensions
of the panel, N and T, and maybe some notion of what the data are like
(quantitative or dummies, strictly positive or not...)?
Allin
>> -----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, August 18, 2015 10:01 AM
>> To: Gretl development
>> Subject: [Gretl-devel] next release?
>>
>> I realize that in the middle of August we're in down-time for many
>> people, so recent gretl snapshots probably haven't had much of a workout.
>>
>> All the same, let me ask: has anyone found problems lately that
>> should be addressed before we release gretl 1.10.2? I'd like to do so
>> before the semester starts here (i.e. by the end of next week).
_______________________________________________
Gretl-devel mailing list
Gretl-devel(a)lists.wfu.edu
http://lists.wfu.edu/mailman/listinfo/gretl-devel
12:54 p.m.
On Tue, 18 Aug 2015, Summers, Peter wrote:
(re. panel plot failing in the case of multiple plots in grid)
> Thanks, Peter. I'll take a look at this. Can you give me an
idea of
> the dimensions of the panel, N and T, and maybe some notion of what
> the data are like (quantitative or dummies, strictly positive or
> not...)?
N = 14, T = 139 (unbalanced). Both variables are quantitative and
non-negative. Below are simple summary stats for the problematic
variable, by country (unit). There's one that has no valid
observations -- could that be the problem?
[...]
Yes, indeed it could. I've now rejigged the multiple plot code to skip
any units that have no valid ovbservations.
Allin
1:38 p.m.
Everything works fine now, Allin. Many thanks as always!!
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, August 18, 2015 12:54 PM
To: Gretl development
Subject: Re: [Gretl-devel] next release?
On Tue, 18 Aug 2015, Summers, Peter wrote:
(re. panel plot failing in the case of multiple plots in grid)
>> Thanks, Peter. I'll take a look at this. Can you give me an idea of
>> the dimensions of the panel, N and T, and maybe some notion of what
>> the data are like (quantitative or dummies, strictly positive or
>> not...)?
>
> N = 14, T = 139 (unbalanced). Both variables are quantitative and
> non-negative. Below are simple summary stats for the problematic
> variable, by country (unit). There's one that has no valid
> observations -- could that be the problem?
[...]
Yes, indeed it could. I've now rejigged the multiple plot code to skip any units
that have no valid ovbservations.
Allin
_______________________________________________
Gretl-devel mailing list
Gretl-devel(a)lists.wfu.edu
http://lists.wfu.edu/mailman/listinfo/gretl-devel
3379
days inactive
3379
days old
5 comments
2 participants
participants (2)
-
Allin Cottrell -
Summers, Peter