2:12 p.m.
Dear Allin,
I think there's something "strange" with the special
accessor $["roots"]. Please take a look at the following
Hansl code:
<hansl>
# Gretl doesn't save the "r" matrix
open data9-7
loop i = 1..2
arima $i 1 2 ; QNC
matrix r = $["roots"] # overwrites "r"
print r
endloop
# Gretl saves the "r" matrix
open data9-7
arima 1 1 2 ; QNC
matrix r = $["roots"]
print r
arima 2 1 2 ; QNC
matrix r = $["roots"] # overwrites "r"
print r
</hansl>
I think this is a little bit weird because I can use the
same ideia with another kind of accessor:
<hansl>
open data9-7
loop i = 1..2
arima $i 1 2 ; QNC
matrix c = $coeff
print c
endloop
open data9-7
loop i = 1..2
arima $i 1 2 ; QNC
matrix t = $test
print t
endloop
open data9-7
loop i = 1..2
arima $i 1 2 ; QNC
matrix p = $pvalue
print p
endloop
</hansl>
Thanks in advance,
Henrique Andrade*
*
7:46 p.m.
On Fri, 25 Jan 2013, Henrique Andrade wrote:
I think there's something "strange" with the special
accessor $["roots"]. Please take a look at the following
Hansl code:
<hansl>
# Gretl doesn't save the "r" matrix
open data9-7
loop i = 1..2
arima $i 1 2 ; QNC
matrix r = $["roots"] # overwrites "r"
print r
endloop
Yes, that sort of thing ought to work in loops, but it didn't.
This should now be fixed in CVS and snapshots.
Allin Cottrell
5:29 p.m.
Em 25 de janeiro de 2013, Allin escreveu:
On Fri, 25 Jan 2013, Henrique Andrade wrote:
> I think there's something "strange" with the special
> accessor $["roots"]. Please take a look at the following
> Hansl code:
>
> <hansl>
> # Gretl doesn't save the "r" matrix
> open data9-7
> loop i = 1..2
> arima $i 1 2 ; QNC
> matrix r = $["roots"] # overwrites "r"
> print r
> endloop
Yes, that sort of thing ought to work in loops, but it didn't.
This should now be fixed in CVS and snapshots.
Now it is working fine. Thanks Allin!
Best,
Henrique Andrade*
*
11:38 a.m.
Em 27 de janeiro de 2013, Henrique escreveu:
Em 25 de janeiro de 2013, Allin escreveu:
On Fri, 25 Jan 2013, Henrique Andrade wrote:
>
> > I think there's something "strange" with the special
> > accessor $["roots"]. Please take a look at the following
> > Hansl code:
> >
> > <hansl>
> > # Gretl doesn't save the "r" matrix
> > open data9-7
> > loop i = 1..2
> > arima $i 1 2 ; QNC
> > matrix r = $["roots"] # overwrites "r"
> > print r
> > endloop
>
> Yes, that sort of thing ought to work in loops, but it didn't.
> This should now be fixed in CVS and snapshots.
>
Now it is working fine. Thanks Allin!
Now it is not working anymore :-( Using the "test case" I provided
everything works:
<hansl>
open data9-7
loop i = 1..2
arima $i 1 2 ; QNC
matrix r = $["roots"] # overwrites "r"
print r
endloop
</hansl>
But with my "real world" example it doesn't:
<hansl>
open australia.gdt
logs E
ldiff E
smpl --no-missing *E
series Y = l_E
scalar max_ar_lags = 2
scalar max_ma_lags = 1
scalar dummies = 1
string Nome = "l_E"
list Projecoes = null
matrix especificacoes = {}
loop for P=0..max_ar_lags --quiet
loop for Q=0..max_ma_lags --quiet
matrix especificacoes = especificacoes | {$P, 1, $Q}
endloop
endloop
scalar linhas = rows(especificacoes)
loop j=1..linhas --quiet
scalar s_ar = especificacoes[j,1]
scalar s_i = especificacoes[j,2]
scalar s_ma = especificacoes[j,3]
if dummies = 1
arima s_ar s_i s_ma ; Y --quiet
sprintf d_up "d_up_%d_%d_%d", s_ar, s_i, s_ma
sprintf d_down "d_down_%d_%d_%d", s_ar, s_i, s_ma
series @d_up = misszero(($uhat >= +2*sd($uhat)))
series @d_down = misszero(($uhat <= -2*sd($uhat)))
arima s_ar s_i s_ma ; Y @d_up @d_down #--quiet
matrix z = $coeff./$stderr
matrix roots = $["roots"]
if abs(z)>critical(t, $T, 0.025) && abs(roots[,1])>1
sprintf Y_hat "Y_hat_%d_%d_%d", s_ar, s_i, s_ma
sprintf Nome_modelo "Arima(%d,%d,%d)", s_ar, s_i, s_ma
fcast @Y_hat
list Projecoes += @Y_hat
setinfo @Y_hat --description="Variável '@Nome' estimada via
modelo '@Nome_modelo'"
endif
else
arima s_ar s_i s_ma ; Y --quiet
endif
endloop
</hansl>
I really can not find what is going wrong :-(
Best regards,
Henrique Andrade*
*
6:10 p.m.
On Wed, 30 Jan 2013, Henrique Andrade wrote:
Em 27 de janeiro de 2013, Henrique escreveu:
Em 25 de janeiro de 2013, Allin escreveu:
>
> On Fri, 25 Jan 2013, Henrique Andrade wrote:
>>
>>> I think there's something "strange" with the special
>>> accessor $["roots"]. Please take a look at the following
>>> Hansl code:
>>>
>>> <hansl>
>>> # Gretl doesn't save the "r" matrix
>>> open data9-7
>>> loop i = 1..2
>>> arima $i 1 2 ; QNC
>>> matrix r = $["roots"] # overwrites "r"
>>> print r
>>> endloop
>>
>> Yes, that sort of thing ought to work in loops, but it didn't.
>> This should now be fixed in CVS and snapshots.
>>
>
> Now it is working fine. Thanks Allin!
Now it is not working anymore :-( Using the "test case" I provided
everything works:
<hansl>
open data9-7
loop i = 1..2
arima $i 1 2 ; QNC
matrix r = $["roots"] # overwrites "r"
print r
endloop
</hansl>
But with my "real world" example it doesn't [...]
loop for P=0..max_ar_lags --quiet
loop for Q=0..max_ma_lags --quiet
matrix especificacoes = especificacoes | {$P, 1, $Q}
endloop
endloop
The trouble is that the first model estimated in your "real
world" example doesn't have any ARMA roots (P = Q = 0). For
all the other specifications you'll get the $["roots"] matrix
OK.
Allin Cottrell
6:59 p.m.
Em 30 de janeiro de 2013, Allin escreveu:
On Wed, 30 Jan 2013, Henrique Andrade wrote:
(...)
> Now it is not working anymore :-( Using the "test case" I provided
> everything works:
>
> <hansl>
> open data9-7
> loop i = 1..2
> arima $i 1 2 ; QNC
> matrix r = $["roots"] # overwrites "r"
> print r
> endloop
> </hansl>
>
> But with my "real world" example it doesn't [...]
> loop for P=0..max_ar_lags --quiet
> loop for Q=0..max_ma_lags --quiet
> matrix especificacoes = especificacoes | {$P, 1, $Q}
> endloop
> endloop
The trouble is that the first model estimated in your "real
world" example doesn't have any ARMA roots (P = Q = 0). For
all the other specifications you'll get the $["roots"] matrix
OK.
Ok, I have to recognize: I need to change my real world :-(
Thanks for your help and sorry for my poor real world ;-)
--
Henrique Andrade*
*
4343
days inactive
4348
days old
5 comments
2 participants
participants (2)
-
Allin Cottrell -
Henrique Andrade