11:42 a.m.
Hi,
I would like to report on the feedback that I received during the two
42 hr. applied econometrics with gretl trainings that I have given
over the last 3 weeks.
In general, the students (almost all of whom were professionals from
the government) did very much enjoy learning and using gretl. They
also understood the development paradigm and how it is possible that
gretl is free yet as good as various commercial programs. I am sure
that in the future they will be using gretl extensively, although they
will continue to use other programs mainly SPSS, Stata and Eviews.
There were also 3 issues that they raised:
1)- Some people were very much interested in the principal components
functionality but they were ultimately disappointed. Apparently they
need to use this functionality (in SPSS) to create various indexes
using a large number of series. Still, I guess this can be considered
more like a statistical functionality than an econometric one, no?
2)- Some students suggested (and all others agreed) that it would be
very useful to have a predict command, which will provide predicted
values as well as slopes (given Xs) for various nonlinear models such
as polynomial regressions, logit, probit etc. I think this could be
nice to have as a command as well as a GUI entry next to the forecast
item. Maybe a small goodie to consider for the 2.0 release? They said
Stata has this.
3)- One student told that the "save to session as icon" and the "save
as icon and close" items act in a confusing manner. The suggestion was
that:
a)- Once the user selects "save to session as icon", this entry
becomes ghosted so that it cannot be selected again to get an
unnecessary error message.
b)- If the "save to session as icon" is already chosen, the "save
as icon and close" does not generate an error message but just close
the model window, which is already saved.
I hope the above feedback is useful. I will probably give more
trainings over the next months and I will let you know if I get more
comments.
Cheers
Talha
--
“An expert is a person who has made all the mistakes that can be made
in a very narrow field.” - Niels Bohr (1885-1962)
--
8:19 p.m.
On Mon, 27 Feb 2012, Talha Yalta wrote:
I would like to report on the feedback that I received during the
two
42 hr. applied econometrics with gretl trainings that I have given
over the last 3 weeks.
In general, the students (almost all of whom were professionals from
the government) did very much enjoy learning and using gretl. They
also understood the development paradigm and how it is possible that
gretl is free yet as good as various commercial programs. I am sure
that in the future they will be using gretl extensively, although they
will continue to use other programs mainly SPSS, Stata and Eviews.
There were also 3 issues that they raised:
1)- Some people were very much interested in the principal components
functionality but they were ultimately disappointed. Apparently they
need to use this functionality (in SPSS) to create various indexes
using a large number of series. Still, I guess this can be considered
more like a statistical functionality than an econometric one, no?
What in particular didn't they like about gretl's "pca"
command (and/or the princomp() function)? What's missing from
their point of view?
2)- Some students suggested (and all others agreed) that it would be
very useful to have a predict command, which will provide predicted
values as well as slopes (given Xs) for various nonlinear models such
as polynomial regressions, logit, probit etc. I think this could be
nice to have as a command as well as a GUI entry next to the forecast
item. Maybe a small goodie to consider for the 2.0 release? They said
Stata has this.
I don't see what the difference is between "predicted values"
and what we offer already (in sample fitted values and
out-of-sample forecasts). Can you expand on what you mean?
3)- One student told that the "save to session as icon" and
the "save
as icon and close" items act in a confusing manner. The suggestion was
that:
a)- Once the user selects "save to session as icon", this entry
becomes ghosted so that it cannot be selected again to get an
unnecessary error message.
Yes, that's obviously right. Fixed in CVS and snapshots.
b)- If the "save to session as icon" is already chosen,
the "save
as icon and close" does not generate an error message but just close
the model window, which is already saved.
I think that one should be disabled too, leaving the plain
"Close" option.
Allin
7:25 a.m.
On Tue, Feb 28, 2012 at 10:19 PM, Allin Cottrell <cottrell(a)wfu.edu> wrote:
On Mon, 27 Feb 2012, Talha Yalta wrote:
> 1)- Some people were very much interested in the principal components
> functionality but they were ultimately disappointed. Apparently they
> need to use this functionality (in SPSS) to create various indexes
> using a large number of series. Still, I guess this can be considered
> more like a statistical functionality than an econometric one, no?
What in particular didn't they like about gretl's "pca"
command (and/or the princomp() function)? What's missing from
their point of view?
These people were from State Planning Agency. They told me
that they
have about 60 series (which have different levels of collinearity) and
they use SPSS to do principal components analysis to create regional
development indices (Turkey has 81 provinces). I am not very familiar
with pca but I can call them and learn more.
> 2)- Some students suggested (and all others agreed) that it would
be
> very useful to have a predict command, which will provide predicted
> values as well as slopes (given Xs) for various nonlinear models such
> as polynomial regressions, logit, probit etc. I think this could be
> nice to have as a command as well as a GUI entry next to the forecast
> item. Maybe a small goodie to consider for the 2.0 release? They said
> Stata has this.
I don't see what the difference is between "predicted values"
and what we offer already (in sample fitted values and
out-of-sample forecasts). Can you expand on what you mean?
Now this is maybe I
didn't know how to fully use gretl in this
context. The issue arised on 2 occasions:
(1) I had a polynomial regression and I was showing them to enter from
the GUI the command something like:
prediction = $coeff[1] + $coeff[2]*x + $coeff[3]*x^2
(2) I was showing an ordered logit example and I had long commands like:
pcut0 = 1 / (1+exp(-$coeff[1]-x*$coeff[2])+exp(-$coeff[1]-x*$coeff[3]))
pcut1= exp(-$coeff[1]-x*$coeff[2]) /
(1+exp(-$coeff[1]-x*$coeff[2])+exp(-$coeff[1]-x*$coeff[3]))
pcut2= exp(-$coeff[1]-x*$coeff[3]) /
(1+exp(-$coeff[1]-x*$coeff[2])+exp(-$coeff[1]-x*$coeff[3]))
...and they said Stata (supposedly) has a command where you enter x
and get the prediction and slope for different models :-P
I think that one should be disabled too, leaving the plain
"Close" option.
Yes this is probably the best. Thanks for the fix.
Sincerely
Talha
--
“An expert is a person who has made all the mistakes that can be made
in a very narrow field.” - Niels Bohr (1885-1962)
--
4702
days inactive
4704
days old
2 comments
2 participants
participants (2)
-
Allin Cottrell -
Talha Yalta