gnuplot: set pointtype
by artur tarassow

I would like to set the point type for a simple scatter plot, but it does
not work. I've got something like:
gnuplot XY --output=display { set pointtype 1 ; }
Instead, the option "pointsize" works, but not pointtype.
Is there a simple workaround available?
Thanks!
Artur
6 years, 9 months

boxplot command flexibility
by artur tarassow

Dear gretl programmers,
at the moment it is not possible to set further plotting options using the
boxplot-command.
Would it possible to allow more flexibility for the user such as
"{ set title '' ; }" or "set grid" "set ylabel" etc.?
Best,
Artur
6 years, 9 months

vector autoregressive
by aymen kaabi

good evening
i have a problem with gretl_var.
first with the function ar_model i can specify the lags
FOR EXAMPLE: Yt= Yt-3 + Y t-2 + Yt-6 +CONST but with gretl I CAN'T i
just enter the lag order .
HOW can i specify the lags .
my project work personalized lags please answer me .
thanks for all
6 years, 10 months

error message: cURL error 7 (Couldn't connect to server)
by artur tarassow

Hi all,
for some days now, I cannot access the function-package server anymore.
This happens both on Linux and Windows (using current cvs). The error
message I obtain is:
"cURL error 7 (Couldn't connect to server)"
Artur
6 years, 10 months

Coeffiicent Wald Tests across equations for individual coefficents (within SUR)
by Jan Tille

Dear gretl users,
I have a questions regrding coefficient wald tests.
I have estimated a SUR system (20 equations) and now I would like to conduct Wald tests for individual coefficients across different equations.
I know that I can use the restrict command e.g.
restrict "System"
b[1,2]-b[3,2]=0
end restrict
estimate "System" method=sur
which yields an F-test at the end.
However, I want to conduct the tests for several equations (14 out of 20) each with 8 coefficients, which would require a lot of individual restricitons.
Therefore, I asked myself, whether I could obtain the same results using
(b[1,2]-b[3,2])/S(diff),
where S(diff) =sqrt(S(b[1,2])^2+S(b[1,2])^2)
Unfortunately, the results between the two tests differ (even if I square the t-statistic).
Finally, I am looking for a way, to calculate the wald tests, without manually respecifying the every single restriction.
Thanks in advance.
Jan
6 years, 10 months

Magnitude of VAR shocks
by Gabriela Nodari

Dear Gretl Users,
I was wondering if there is the possibility to change the magnitude of a
given structural shock within a SVAR framework. By default, Gretl calculate
IRFs based on a one-standard deviation shock. Can I change this magnitude?
Thanks,
Gabriela
6 years, 10 months

ADF by OLS manually
by yinung＠Gmail

Dear All,
I occasionally tried the following script to run ADF OLS manually. However,
it is found the commands for DF and ADF1 as follow are ok but not for ADF2.
<hansl>
open data3-6
#ADF by OLS manually
DF <- ols diff(Yt) const Yt(-1)
ADF1 <- ols diff(Yt) const Yt(-1) diff(Yt_1)
ADF2 <- ols diff(Yt) const Yt(-1) diff(Yt_1) diff(Yt_2)
</hansl>
I know that it formally should be execute before these ols commands:
lags 2; Yt
But this looks like Yt_1 has been generated internally since the script
like this does not work:
ADF1 <- ols diff(Yt) const Yt(-1) diff(Y(-1))
I am not asking any revision. Just curious. Thanks.
Yi-Nung Yang
6 years, 10 months

Gretl Help
by Antti Erkkison

Hi, sorry for this e-mail, but I have not found any other way to find this
answer.
I am an economics student from Finland that are using Gretl, and finding it
useful tool to analyze data.
We were wondering if Gretl had a function for dividing groups into low and
high respondents, in other word we want to test a part of a population (in
OLS) who have marked the highest scores in a survey. We want to test this
against other answers in the same survey, but are in doubt that Gretl
understands that person 1 is the same person when we change the order of
respondents by sorting them in the dependent value and then setting the
range to half of the population .
I would really appreciate an answer, thank you.
Best regards
Antti Erkkison
6 years, 10 months

A time series plot produces an error
by yinung＠Gmail

Dear All,
I use the attached file: ex-IBM-2004-09.gdt. It then generated an error (as
shown below) when I tried to have a time-series plot with any variables in
that file.
I tried other "daily data" like djclose.gdt but no such error.
This error can be replicated in gretl 1.9.12 and 1.9.12cvs.
I just tried the version of 1.9.11 (under winXP). It seemed OK with the
same time series plotting.
Thanks
Yi-Nung Yang
===== error message =====
set xrange [1.797693135e+308:1.797693135e+308]
^
"/home/yinung/.gretl/gpttmp.05Vd0T", line 19: undefined value
6 years, 10 months