3:19 a.m.
On Thu, 10 Feb 2011, Summers, Peter wrote:
While helping a student sort out a dynamic panel estimation, I
discovered the following. If I take a data set that's organized
as stacked time series, change the structure to stacked
cross-sections, then change it back, the second change isn't
implemented. I get a box saying no changes were made. Is this
intended behavior?
Ah, maybe it's a bit confusing but in fact it is intended. The
main point is that a panel dataset _must_ be organized as stacked
time series for use in gretl. "Stacked cross sections" is not an
option for panel data in gretl, it's just a way of saying that
your data are currently the wrong way round and need to be fixed.
So, if you go to "/Data/Dataset structure" and say that your
dataset is stacked cross sections, gretl will reorganize it for
you as stacked time series (a "physical" reorganization of the
data). Note that if you then go back to the "Dataset structure"
dialog your data will initially appear as "Stacked time series",
so, naturally, stating that the structure is stacked time series
will produce no change.
If you want to change the actual data layout back to what it was
originally, you have to say that it's stacked cross sections
(again), and gretl will reorganize the data in the opposite
direction.
Allin Cottrell
4:17 p.m.
New subject: changing panel data structure
Am 11.02.2011 04:19, schrieb Allin Cottrell:
On Thu, 10 Feb 2011, Summers, Peter wrote:
> While helping a student sort out a dynamic panel estimation, I
> discovered the following. If I take a data set that's organized
> as stacked time series, change the structure to stacked
> cross-sections, then change it back, the second change isn't
> implemented. I get a box saying no changes were made. Is this
> intended behavior?
Ah, maybe it's a bit confusing but in fact it is intended. The
main point is that a panel dataset _must_ be organized as stacked
time series for use in gretl. "Stacked cross sections" is not an
option for panel data in gretl, it's just a way of saying that
your data are currently the wrong way round and need to be fixed.
So, if you go to "/Data/Dataset structure" and say that your
dataset is stacked cross sections, gretl will reorganize it for
you as stacked time series (a "physical" reorganization of the
data). Note that if you then go back to the "Dataset structure"
dialog your data will initially appear as "Stacked time series",
so, naturally, stating that the structure is stacked time series
will produce no change.
If you want to change the actual data layout back to what it was
originally, you have to say that it's stacked cross sections
(again), and gretl will reorganize the data in the opposite
direction.
Allin Cottrell
Hi there,
in terms of changing the layout back from the stacked-cross-section
format to the original -stacked-time-series shape I found the following
behaviour:
1. starting with
index
CN:1 1
CN:2 2
CN:3 3
GER:1 4
GER:2 5
GER:3 6
2. performing setobs T 1.1 --stacked-cross-section yields the
intended format
index
CN:1 1
GER:1 4
CN:2 2
GER:2 5
CN:3 3
GER:3 6
3. in order to reshape the data back to the organization of (1.) the command
setobs T 1:01 --stacked-time-series does not alter the oganization
of the dataset.
Instead one has to interate the command setobs T 1.1
--stacked-cross-section.
After 3 steps the result equals the one of 1.
But perhaps I mis-specified the frequency statement(s).
Best
Pindar
_______________________________________________
Gretl-users mailing list
Gretl-users(a)lists.wfu.edu
http://lists.wfu.edu/mailman/listinfo/gretl-users
12:31 p.m.
New subject: Sign mix-up for ordered logit
Hello all,
Coefficients from the Ordered Logit function are correct in gretl, but
the signs seem to be mixed up.
According to the user guide, gretl fits an ascending cumulative odds
model. Then it predicts the odds of being at or below a particular
category. In this case the slopes for independent variables simply have
wrong signs.
To make the model more similar to OLS and common sense, descending
models could be fitted. (Aggregating at or beyond). Then all the cut
points have wrong signs now, and the coefficients are correct.
I cannot see that this is mentioned in change logs or discussion lists.
My gretl version is 1.8.7 on Ubuntu 10.04.
Am I correct here, or am I missing something? (If needed I can send R
comparisons and spreadsheets converting to probabilities for models in
the book "Logistic Regression Models for Ordinal Response Variables" by
Ann A. O'Connell).
Best regards
Birger
5 p.m.
New subject: Sign mix-up for ordered logit
On Sat, 12 Feb 2011, Birger Baksaas wrote:
Coefficients from the Ordered Logit function are correct in gretl,
but
the signs seem to be mixed up.
[...]
If needed I can send R comparisons and spreadsheets converting to
probabilities for models in the book "Logistic Regression Models for
Ordinal Response Variables" by Ann A. O'Connell.
Please do. That would be very helpful.
Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche
r.lucchetti(a)univpm.it
http://www.econ.univpm.it/lucchetti
4:59 p.m.
New subject: changing panel data structure
Allin,
Thanks for the clarifications. One thing though -- if I change the data structure to
stacked cross sections, the data get re-organized as expected. But if I then say it's
stacked cross sections a second time, the second re-organization doesn't restore the
original stacked time series.
Here's a sample of the data:
1) original (correct) stacked time series:
Argentina 14244.71
Argentina 12275.32
Argentina 11118.07
Argentina 13661.23
Argentina 14503.41
Argentina 15670.38
Australia 22988.74
Australia 24901.34
Australia 26079.21
Australia 29283.26
Australia 33732.40
Australia 38077.99
Austria 23277.40
Austria 24792.66
Austria 28504.44
Austria 30264.96
Austria 34459.28
Austria 36360.78
2) after applying stacked cross-sections:
Argentina 14244.71
Colombia 8057.55
Greece 19121.09
Korea, Republic 18670.63
Pakistan 2273.91
Sri Lanka 4455.84
Argentina 12275.32
Colombia 7888.27
Greece 18317.34
Korea, Republic 20788.10
Pakistan 2673.16
Sri Lanka 5290.79
Argentina 11118.07
Colombia 8416.99
Greece 18804.01
Korea, Republic 24389.44
Pakistan 3102.81
Sri Lanka 6081.70
Argentina 13661.23
3) after 2nd application of stacked cross-sections:
Argentina 14244.71
Malawi 1255.76
Belgium 22800.87
Mexico 12464.08
Burkina Faso 1218.20
Nicaragua 2595.55
Colombia 8057.55
Paraguay 6163.80
Denmark 33573.49
Rwanda 1351.55
El Salvador 6470.20
South Africa 9856.35
Greece 19121.09
Syria 2878.26
Indonesia 3094.67
Turkey 7561.41
Jamaica 9418.74
Venezuela 13117.04
Korea, Republic 18670.63
PS
-----Original Message-----
From: gretl-users-bounces(a)lists.wfu.edu [mailto:gretl-users-bounces@lists.wfu.edu] On
Behalf Of Allin Cottrell
Sent: Thursday, February 10, 2011 9:19 PM
To: Gretl list
Subject: Re: [Gretl-users] changing panel data structure
On Thu, 10 Feb 2011, Summers, Peter wrote:
While helping a student sort out a dynamic panel estimation, I
discovered the following. If I take a data set that's organized
as stacked time series, change the structure to stacked
cross-sections, then change it back, the second change isn't
implemented. I get a box saying no changes were made. Is this
intended behavior?
Ah, maybe it's a bit confusing but in fact it is intended. The
main point is that a panel dataset _must_ be organized as stacked
time series for use in gretl. "Stacked cross sections" is not an
option for panel data in gretl, it's just a way of saying that
your data are currently the wrong way round and need to be fixed.
So, if you go to "/Data/Dataset structure" and say that your
dataset is stacked cross sections, gretl will reorganize it for
you as stacked time series (a "physical" reorganization of the
data). Note that if you then go back to the "Dataset structure"
dialog your data will initially appear as "Stacked time series",
so, naturally, stating that the structure is stacked time series
will produce no change.
If you want to change the actual data layout back to what it was
originally, you have to say that it's stacked cross sections
(again), and gretl will reorganize the data in the opposite
direction.
Allin Cottrell
_______________________________________________
Gretl-users mailing list
Gretl-users(a)lists.wfu.edu
http://lists.wfu.edu/mailman/listinfo/gretl-users
5063
days inactive
5064
days old
4 comments
5 participants
participants (5)
-
Allin Cottrell -
Birger Baksaas -
Leon Unger -
Riccardo (Jack) Lucchetti -
Summers, Peter