9:10 a.m.
Hello I am a new gretl user and have a question whether and how it is possible to run a
loop only via a selection of time series that have minimum length of 24 months.
The series's have different start and end dates and I want to run a regression for
each series that has a minimum of 24 data points.
Is there anybody, who can hint me on how to restrict or define the loop properly. On
possibility is, to preselect the data that I import into gretl but this would be tedious,
because I will also run a different model on all series that have a minimum of 12 data
point.
Thanks in advance,
Jan
10:12 a.m.
Am 06.07.2012 11:10, schrieb Jan Tille:
Hello I am a new gretl user and have a question whether and how it
is
possible to run a loop only via a selection of time series that have
minimum length of 24 months. The series's have different start and
end dates and I want to run a regression for each series that has a
minimum of 24 data points. Is there anybody, who can hint me on how
to restrict or define the loop properly. On possibility is, to
preselect the data that I import into gretl but this would be
tedious, because I will also run a different model on all series that
have a minimum of 12 data point.
Thanks in advance,
Jan
Hi Jan,
one easy way would be to check the observation numbers used in each
regression by looking at the value of the accessor variable $T, like so:
<hansl>
ols y const x
if $T>=24
# do stuff
else
# discard the results
endif
</hansl>
If running the redundant regressions is too costly computationally
(which I very much doubt given the small number of observations
involved), you could check the sample properties based on the given
variable (y) before doing anything:
<hansl>
smpl --no-missing y
if $nobs>= 24
ols y const x
# do stuff
else
# do not do anything
endif
</hansl>
HTH,
sven
11:44 a.m.
Sven, thank you very much. I will try it. I tried $nobs, but $nobs always returned 124,
which is the length of the total sample period.
Regards,
Jan
________________________________________
Von: Plus.line MailSystem [cyrus(a)mailer.plusline.de] im Auftrag von Sven Schreiber
[svetosch(a)gmx.net]
Gesendet: Freitag, 6. Juli 2012 12:12
An: gretl-users(a)lists.wfu.edu
Betreff: Re: [Gretl-users] Restricting a sample to specific time series
Am 06.07.2012 11:10, schrieb Jan Tille:
Hello I am a new gretl user and have a question whether and how it
is
possible to run a loop only via a selection of time series that have
minimum length of 24 months. The series's have different start and
end dates and I want to run a regression for each series that has a
minimum of 24 data points. Is there anybody, who can hint me on how
to restrict or define the loop properly. On possibility is, to
preselect the data that I import into gretl but this would be
tedious, because I will also run a different model on all series that
have a minimum of 12 data point.
Thanks in advance,
Jan
Hi Jan,
one easy way would be to check the observation numbers used in each
regression by looking at the value of the accessor variable $T, like so:
<hansl>
ols y const x
if $T>=24
# do stuff
else
# discard the results
endif
</hansl>
If running the redundant regressions is too costly computationally
(which I very much doubt given the small number of observations
involved), you could check the sample properties based on the given
variable (y) before doing anything:
<hansl>
smpl --no-missing y
if $nobs>= 24
ols y const x
# do stuff
else
# do not do anything
endif
</hansl>
HTH,
sven
_______________________________________________
Gretl-users mailing list
Gretl-users(a)lists.wfu.edu
http://lists.wfu.edu/mailman/listinfo/gretl-users
2:02 p.m.
On 07/09/2012 01:44 PM, Jan Tille wrote:
Sven, thank you very much. I will try it. I tried $nobs, but $nobs
always returned 124, which is the length of the total sample period.
No, $nobs should be related to the currently selected sample, and thus
after doing
smpl --no-missing y
it should be different.
-sven
11:44 a.m.
okay, but I have times series (fund returns) with different start as well as end dates and
therefore missing values might occur at the start at the end or both. if I restrict my
sample to the ones with no missing values, I guess that I will come up with funds that
have a complete history.
kind regards,
Jan
________________________________________
Von: Plus.line MailSystem [cyrus(a)mailer.plusline.de] im Auftrag von Sven Schreiber
[svetosch(a)gmx.net]
Gesendet: Montag, 9. Juli 2012 16:02
An: gretl-users(a)lists.wfu.edu
Betreff: Re: [Gretl-users] Restricting a sample to specific time series
On 07/09/2012 01:44 PM, Jan Tille wrote:
Sven, thank you very much. I will try it. I tried $nobs, but $nobs
always returned 124, which is the length of the total sample period.
No, $nobs should be related to the currently selected sample, and thus
after doing
smpl --no-missing y
it should be different.
-sven
_______________________________________________
Gretl-users mailing list
Gretl-users(a)lists.wfu.edu
http://lists.wfu.edu/mailman/listinfo/gretl-users
10:34 a.m.
On Fri, 6 Jul 2012, Jan Tille wrote:
Hello I am a new gretl user and have a question whether and how it is
possible to run a loop only via a selection of time series that have
minimum length of 24 months. The series's have different start and end
dates and I want to run a regression for each series that has a minimum
of 24 data points. Is there anybody, who can hint me on how to restrict
or define the loop properly. On possibility is, to preselect the data
that I import into gretl but this would be tedious, because I will also
run a different model on all series that have a minimum of 12 data
point.
Maybe you can take a couple of ideas from the script below:
<hansl>
open AWM.gdt --quiet
smpl 1994:1 ;
loop foreach i EEN EEN_DIS EER FDD GCD GCN --quiet
scalar n = sum(ok($i))
printf "$i: %d\n", n
if n>16
printf "OK!\n"
else
printf "Oh, bother!\n"
endif
end loop
</hansl>
--------------------------------------------------
Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche
(formerly known as Università di Ancona)
r.lucchetti(a)univpm.it
http://www2.econ.univpm.it/servizi/hpp/lucchetti
--------------------------------------------------
11:46 a.m.
Riccardo, thank you for the script. I will read it in detail and try it out.
Kind regards,
Jan
________________________________________
Von: Plus.line MailSystem [cyrus(a)mailer.plusline.de] im Auftrag von Riccardo (Jack)
Lucchetti [r.lucchetti(a)univpm.it]
Gesendet: Freitag, 6. Juli 2012 12:34
An: Gretl list
Betreff: Re: [Gretl-users] Restricting a sample to specific time series
On Fri, 6 Jul 2012, Jan Tille wrote:
Hello I am a new gretl user and have a question whether and how it
is
possible to run a loop only via a selection of time series that have
minimum length of 24 months. The series's have different start and end
dates and I want to run a regression for each series that has a minimum
of 24 data points. Is there anybody, who can hint me on how to restrict
or define the loop properly. On possibility is, to preselect the data
that I import into gretl but this would be tedious, because I will also
run a different model on all series that have a minimum of 12 data
point.
Maybe you can take a couple of ideas from the script below:
<hansl>
open AWM.gdt --quiet
smpl 1994:1 ;
loop foreach i EEN EEN_DIS EER FDD GCD GCN --quiet
scalar n = sum(ok($i))
printf "$i: %d\n", n
if n>16
printf "OK!\n"
else
printf "Oh, bother!\n"
endif
end loop
</hansl>
--------------------------------------------------
Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche
(formerly known as Università di Ancona)
r.lucchetti(a)univpm.it
http://www2.econ.univpm.it/servizi/hpp/lucchetti
--------------------------------------------------
4547
days inactive
4551
days old
6 comments
3 participants
participants (3)
-
Jan Tille -
Riccardo (Jack) Lucchetti -
Sven Schreiber