On 16/02/13 17:01, Sven Schreiber wrote:
Am 16.02.2013 16:21, schrieb Allin Cottrell:

On Fri, 15 Feb 2013, Ignacio Diaz-Emparanza wrote:


BTW, when you ask gretl to add "periodic dummies" in such a way the names of 
the dummy variables depend on the seasonal periodicity, i.e. if the data are 
quarterly the dummies are dq1,..., dq4, if the data are monthly they are 
dm1,..., dm12 if the seasonal periodicity is greather than 12 the dummies are 
dummy_1,..., dummy_99 and then dummy100, dummy101, etc

At the risk of introducing another backward incompatible change, I would 
suggest our dear developers that a unique name should be used for the 
seasonal dummies independently on the periodicity. And I propose to use d1, 
d2, ....

I don't have any objection to that, other than the backward
incompatibility. Any others have thoughts on the matter?


Ignacio, could you explain the advantages? Or the other way around, the
disadvantages of having different names for different periodicities?
1) It is easier for the user to have a unique name.

2) For a user who wants to use the seasonal dummies inside a function it is really uncomfortable not knowing beforehand what are their names, so having to use always an ugly intermediate function like this below while changing to a unique name may avoid it:

<hansl>
function list seasdum (void)
    list deterministic = null
    genr dummy
    if ($pd=4)
        list deterministic += dq1 dq2 dq3
    elif ($pd=12)
        list deterministic += dm1 dm2 dm3 dm4 dm5 dm6 dm7 dm8 dm9 dm10 dm11
    else
        scalar spd1=$pd-1
        loop i=1..spd1 --quiet
            if (i<100)
                list deterministic += dummy_$i
            else
                list deterministic += dummy$i
            endif
        endloop
    endif
return deterministic
end function

nulldata 100
#Quarterly data
setobs 4 1:1 --time-series
list sdum=seasdum()
list sdum --print

nulldata 100
#Monthly data
setobs 12 1:1 --time-series
list sdum=seasdum()
list sdum --print

nulldata 1000
#Hourly data
setobs 24 1:1 --time-series
list sdum=seasdum()
list sdum --print


nulldata 1000
#A rather extreme case:
#Weekly periodicity with hourly data
setobs 168 1:1 --time-series
list sdum=seasdum()
list sdum --print
</hansl>

I see now from Jack's response that the function might be made now much simpler using

q=$obsminor
list D=dummify(q)

and if $obsminor is allowed as a parameter of 'dummify' in only one line:

list D=dummify($osbminor)

(But Jacks's example does not work for $pd=168)

 
Actually, if the seasonal dummies business were touched, I would propose
to think also about covering centered seasonal dummies, as gretl already
does in the VECM case.
I also support this.

--
Firma Arista
Ignacio Díaz-Emparanza
Zuzendaria/Director
ignacio.diaz-emparanza@ehu.es
94 6013732
EKONOMIA APLIKATUA III SAILA (EKONOMETRIA ETA ESTATISTIKA)/ DEPARTAMENTO DE ECONOMÍA APLICADA III (ECONOMETRÍA Y ESTADÍSTICA)
UPV/EHU
Avda. Lehendakari Aguirre, 83 | 48015 BILBAO
T.: +34 946013740 | F.: +34 946013754
 www.ea3.ehu.es
ERNE! Baliteke mezu honen zatiren bat edo mezu osoa legez babestuta egotea. Mezuak badu bere hartzailea. Okerreko helbidera heldu bada (helbidea gaizki idatzi, transmisioak huts egin) eman abisu igorleari, korreo honi erantzunda. Kontuz! Mezua ez bada zuretzat, ez erabili, ez zabaldu beste inori, ez kopiatu eta ez baliatu.
¡ATENCIÓN! Este mensaje contiene información privilegiada o confidencial a la que sólo tiene derecho a acceder el destinatario. Si usted lo recibe por error le agradeceríamos que no hiciera uso de la información y que se pusiese en contacto con el remitente.

E-mail hau inprimatu baino lehen egiaztatu inprimatzeko beharra.
Antes de imprimir este e-mail piense bien si es necesario hacerlo.