5:45 a.m.
When I run the attached script, it seems the function is correctly
generating the filtered series, and they are available from the command
line, but they don't appear in the main window, so I cannot make the
graphs from the menus. Some time ago this was different.
This is on Ubuntu linux 12.04 and with gretl 1.9.9cvs downloaded today,
but I also tried with a version from a month ago with the same problem.
So I think the question about the length of the names is not the problem
here.
--
Ignacio Diaz-Emparanza
DEPARTAMENTO DE ECONOMÍA APLICADA III (ECONOMETRÍA Y ESTADÍSTICA)
UPV/EHU Avda. Lehendakari Aguirre, 83 | 48015 BILBAO
T.: +34 946013732 | F.: +34 946013754
www.ehu.es/ignacio.diaz-emparanza/
www.ea3.ehu.es
7:28 a.m.
On Wed, 10 Oct 2012, Ignacio Diaz-Emparanza wrote:
When I run the attached script, it seems the function is correctly
generating
the filtered series, and they are available from the command line, but they
don't appear in the main window, so I cannot make the graphs from the menus.
Some time ago this was different.
This is on Ubuntu linux 12.04 and with gretl 1.9.9cvs downloaded today, but I
also tried with a version from a month ago with the same problem. So I think
the question about the length of the names is not the problem here.
Totally unrelated to your problem, but still helpful, I hope: your
function could be made a bit more compact and possibly slightly more
efficient by using vectors instead of loops where possible:
<hansl>
function list seasfSTB(const series *y)
string yn=argname(y)
#scalar pd=$pd
scalar pd1=$pd-1
scalar fpd2=floor($pd/2)
scalar pd2=$pd/2
scalar s21 = (pd2==fpd2) ? floor($pd/2)-1 : floor($pd/2)
matrix b0 = ones($pd,1)
# all frequencies except zero
series @yn0 = (obs>=$pd) ? filter(y,b0) : NA
if pd2==fpd2
matrix b2 = cos($pi * seq(1,$pd))
series @ynpi = (obs>=$pd) ? filter(y,b2) : NA
list fil = @yn0(-1) @ynpi(-1)
else
list fil = @yn0(-1)
endif
loop j=1..s21 --quiet
scalar w = 2*$pi/$pd * j
matrix frq = seq(1,$pd) * w
series @yna$j=(obs>=$pd) ? -filter(y,cos(frq)) : NA
series @ynb$j=(obs>=$pd) ? filter(y,sin(frq)) : NA
list fil += @yna$j(-1) @ynb$j(-1)
endloop
return fil
end function
</hansl>
Also, you can omit the third argument to filter() if null. Finally, you
may want to start using $pi instead of pi, which will probably be
deprecated in 2.0 (this will only work with CVS, though).
--------------------------------------------------
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
--------------------------------------------------
7:46 a.m.
On 10/10/12 13:28, Riccardo (Jack) Lucchetti wrote:
On Wed, 10 Oct 2012, Ignacio Diaz-Emparanza wrote:
> When I run the attached script, it seems the function is correctly
> generating the filtered series, and they are available from the
> command line, but they don't appear in the main window, so I cannot
> make the graphs from the menus. Some time ago this was different.
>
> This is on Ubuntu linux 12.04 and with gretl 1.9.9cvs downloaded
> today, but I also tried with a version from a month ago with the same
> problem. So I think the question about the length of the names is not
> the problem here.
Totally unrelated to your problem, but still helpful, I hope: your
function could be made a bit more compact and possibly slightly more
efficient by using vectors instead of loops where possible:
Thanks.
It is always convenient to be more efficient :-)
--
Ignacio Diaz-Emparanza
DEPARTAMENTO DE ECONOMÍA APLICADA III (ECONOMETRÍA Y ESTADÍSTICA)
UPV/EHU Avda. Lehendakari Aguirre, 83 | 48015 BILBAO
T.: +34 946013732 | F.: +34 946013754
www.ehu.es/ignacio.diaz-emparanza/
www.ea3.ehu.es
11:59 a.m.
On Wed, 10 Oct 2012, Ignacio Diaz-Emparanza wrote:
When I run the attached script, it seems the function is
correctly generating the filtered series, and they are
available from the command line, but they don't appear in
the main window, so I cannot make the graphs from the menus.
Ah, they're not showing up because they're poor, homeless
orphans.
The series created by your function are lags, and lags are not
normally shown in their own right in the GUI main window;
they're tucked under their parent. But in this case the
parents are deceased, and the lags don't show up at all.
That's now fixed in CVS.
Allin
12:08 p.m.
On Wed, 10 Oct 2012, Allin Cottrell wrote:
On Wed, 10 Oct 2012, Ignacio Diaz-Emparanza wrote:
> When I run the attached script, it seems the function is
> correctly generating the filtered series, and they are
> available from the command line, but they don't appear in
> the main window, so I cannot make the graphs from the menus.
Ah, they're not showing up because they're poor, homeless
orphans.
The series created by your function are lags, and lags are not
normally shown in their own right in the GUI main window;
they're tucked under their parent. But in this case the
parents are deceased, and the lags don't show up at all.
That's now fixed in CVS.
This is the bare minimum you'd expect from an application which bears the
name of a poor girl who was lost in the woods, at the risk of becoming the
succulent dinner of an evil witch.
--------------------------------------------------
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
--------------------------------------------------
12:34 p.m.
On Wed, 10 Oct 2012, Riccardo (Jack) Lucchetti wrote:
On Wed, 10 Oct 2012, Allin Cottrell wrote:
> On Wed, 10 Oct 2012, Ignacio Diaz-Emparanza wrote:
>
>> When I run the attached script, it seems the function is
>> correctly generating the filtered series, and they are
>> available from the command line, but they don't appear in
>> the main window, so I cannot make the graphs from the menus.
>
> Ah, they're not showing up because they're poor, homeless
> orphans.
>
> The series created by your function are lags, and lags are not
> normally shown in their own right in the GUI main window;
> they're tucked under their parent. But in this case the
> parents are deceased, and the lags don't show up at all.
> That's now fixed in CVS.
This is the bare minimum you'd expect from an application which bears the
name of a poor girl who was lost in the woods, at the risk of becoming the
succulent dinner of an evil witch.
;-)
I've now fixed another variant of this "orphan lag" bug,
triggered by the following:
<hansl>
open data9-7
series QNC_1 = QNC(-1)
delete QNC
series QNC = normal()
</hansl>
In this case QNC_1 didn't appear in the GUI. But now it does.
The fix is now in the snapshots too.
Allin
7:56 p.m.
On Wed, 10 Oct 2012, Allin Cottrell wrote:
I've now fixed another variant of this "orphan lag"
bug, triggered by the
following:
<hansl>
open data9-7
series QNC_1 = QNC(-1)
delete QNC
series QNC = normal()
</hansl>
In this case QNC_1 didn't appear in the GUI. But now it does.
The fix is now in the snapshots too.
Let me point out one thing. If you were to run this script
<hansl>
open data9-7
series QNC_1 = QNC(-1)
series QNC = normal()
</hansl>
then QNC_1 would still appear as a "lag child" of series QNC in the
GUI, despite the fact that QNC itself has been redefined, so that
QNC_1 is not the first lag of the new QNC.
This has nothing to do with the issue pointed out by Ignacio
upthread, which was a gretl bug, pure and simple, now fixed. I'm
just saying that there's a limit to the extent that gretl will
attempt to "fix up" perverse user input -- in case anyone was
wondering!
Allin
4421
days inactive
4421
days old
6 comments
3 participants
participants (3)
-
Allin Cottrell -
Ignacio Diaz-Emparanza -
Riccardo (Jack) Lucchetti