Bar graphs
by Talha Yalta
I was teaching someone how to use gretl for research and while showing
the various plot options, the first thing she noticed and asked about
was whether it is possible to have "bar graphs". I think it would
really be useful if we could select "bars" among the various line
types such as points, impulses and steps. This would really open up
for some very nice looking plots involving combinations of lines and
vertical bars especially useful for time-series analysis.
Cheers
Talha
--
"Remember not only to say the right thing in the right place, but far
more difficult still, to leave unsaid the wrong thing at the tempting
moment." - Benjamin Franklin (1706-1790)
--
16 years, 11 months
feature request anniversary
by Sven Schreiber
Dear devel folks,
I just noticed that three days ago an open feature request on the
tracker had its first birthday! Happy birthday to request 1833159, which
by pure coincidence was posted by myself on November 16th, 2007.
But seriously, the request is about "add[ing the] ability to modify an
existing model". The idea was to be able to re-open a saved model and
for example add another lag or perform other modifications, as in other
programs. Essentially this would mean to have a button or menu entry in
the model window, which would open the relevant model specification
dialog with the input fields already filled in with the values of the
saved model.
I think one year is enough time to think about the issue in the sense
that if the actual workers/coders don't like it then we could close the
request as "wontfix" or something like that, or else set some sort of
target, as in which upcoming version the feature would start to appear.
thanks,
sven
17 years
garch dialog
by Sven Schreiber
Today I've noticed (teaching again...) that in the GARCH dialog window
it says "ARCH p: .... ARCH q:". I guess the latter should rather be
"GARCH q", no?
-sven
17 years
Placeholder string
by Cristian Rigamonti
Hi, maybe you have forgotten to replace the string
"note on model statistics abbreviations here"
in lib/src/modelprint.c:2266 with the intended string.
(I've found it in the latest .po file)
Cri
--
GPG/PGP Key-Id 0x943A5F0E - http://www.linux.it/~cri/cri.asc
17 years
Presentation of model results
by Allin Cottrell
Hello all,
We recently redesigned the presentation of coefficients, standard
errors, and so on, in the printing of model results in gretl.
We're now thinking of revamping the presentation of the additional
results that follow the table of coefficients (i.e., sum of
squared residuals, R^2, loglikelihood, etc.).
This has been prompted by Christopher Dougherty, who intends to
use gretl as the standard software in a new edition of his
Introduction to Econometrics (Oxford). He'll be printing a bunch
of regression results and he asked if we'd consider using a more
compact format, along the lines of Eviews.
It's true, our current presentation uses a lot of vertical space.
I've mocked up an alternative compact presentation and I'd like to
hear what people think of this. I'm particularly keen to hear
from translators, because it could be a problem translating the
abbreviated strings.
Please see the example at
http://ricardo.ecn.wfu.edu/~cottrell/dougherty/model.html
where I show a given model as displayed by current gretl, the
proposed modification, and Stata and Eviews output for comparison.
Note that for a translation of the compact table to work
correctly, all the abbreviated strings (such as "Mean dependent
var", "Sum squared resid") should be no greater than 19 characters
in length. Does this look doable?
If translation is not feasible, I'm thinking of including the
compact presentation as an English-only option -- but obviously it
would be preferable to maintain consistency across languages if
possible.
Allin.
17 years
Translation problems
by Henrique
I updated the pt_BR.po files and now I'm testing the translation on my
Macintosh and it works really fine. Additionally, I'm also testing the
translation on Windows computers.
I actually tested in two different Windows based computers and the
results were a little bit confusing (the first display the results
correctly, the other didn't). Please, look the following results:
WINDOWS XP
Média var. dependente 0,004595
D.P. var. dependente 0,041708
Soma resíd. quadrados 0,227919
E.P. da regressão 0,037625
R-quadrado 0,196207
R-quadrado ajustado 0,186222
F(2, 161) 19,65016
P-valor(F) 2,31e-08
Log da verossimilhança 306,7418
Critério de Akaike -607,4836
Critério de Schwarz -598,1840
Critério Hannan-Quinn -603,7083
rô 0,365419
Estatística de Durbin-Watson 1,260998
OS/X Leopard
Média var. dependente 0,004595 D.P. var. dependente 0,041708
Soma resíd. quadrados 0,227919 E.P. da regressão 0,037625
R-quadrado 0,196207 R-quadrado ajustado 0,186222
F(2, 161) 19,65016 P-valor(F) 2,31e-08
Log da verossimilhança 306,7418 Critério de Akaike −607,4836
Critério de Schwarz −598,1840 Critério Hannan-Quinn −603,7083
rô 0,365419 Durbin-Watson 1,260998
I'd realized that the problem is occurring in the line
"Durbin-Watson", that appears "Estatística de Durbin-Watson" in the
Windows XP. I'm a PoEdit user and tried to fix that problem editing
the pt_BR.po, but I didn't found any problems in the .po file.
So, what's the problem? Can you help me to discover what I'm doing wrong?
Best,
Henrique C. de Andrade
Doutorando em Economia Aplicada
Universidade Federal do Rio Grande do Sul
www.ufrgs.br/ppge
P.S.: Allin, I'm speaking about the pt_BR.po file that I`d sent to
you. I though that it was perfectly fine, but when I made the
additional tests I found this problem. I'm sorry!
17 years
bug tracker spam again
by Sven Schreiber
Hi,
it seems that there is a simple pattern to the spam that is posted on
the bug tracker: I get one message in the morning and one in the evening
(GMT+1) every day, all posted into the same tracker item about the HP
filter. Others may get more if it relates to more items.
So my simple forecast would be that this is not going away by itself.
I'm afraid we will have to disable anonymous postings.
cheers (nevertheless...)
sven
17 years
some trivial suggestions
by Sven Schreiber
Hi, these things are just minor issues that I stumbled across:
* when showing the fundamental/descriptive statistics of a variable, the
excess kurtosis is reported. I am now thinking that reporting the
kurtosis itself might be better. The term "excess" is relative to the
implicit benchmark of the Gaussian distribution, but many people argue
that explicit is better than implicit.
* when selecting the frequency distribution of a variable from the menu
you also get a test of whether the normal distribution holds. It would
be nice if the name of the test would be included, given that under
normality tests you get like four different beasts.
* in the histogram graph (via gnuplot) the test result of the normality
test often overlaps with the N(xxxxx,yyyyy) specification details of the
fitted normal curve. I mean literally the letters are overlaid, except
if the fontsize is very small. Maybe the p-value should be put on a
third line instead of behind the test statistic. #
thanks,
sven
17 years
(de.) po question
by Sven Schreiber
I just updated de.po in cvs (after some time without activity), but I
noticed that in one string there appeared something like: '80%'. I was
wondering, does the percentage sign in this context require some escape
character, like '80%%' or '80\%' or whatever?
thanks,
sven
17 years
slight genr problems
by Sven Schreiber
Hi,
on 1.7.9 and current cvs I discovered two minor glitches when I used the
supplied example data sw_ch12.gdt for teaching.
First, when I try to create a new variable referring to the lag of the
price variable PUNEW, but mistakenly change the name to lower case as in
'heyho=punew(-1)', I get an alert dialog window, but with a rather
obscure content, namely:
> genr heyho=punew(
What I would expect is an error message saying something like 'unknown
variable punew'.
Next, when I construct the inflation rate as
'heyho=(PUNEW-PUNEW(-1))/PUNEW(-1)', the variable heyho is listed in the
GUI as if it were the lag of PUNEW, I mean it's not immediately visible,
but the little triangle appears before PUNEW and when you click it you
get to see 'heyho' with the variable ID being indented.
cheers,
sven
17 years, 1 month
