1:47 a.m.
On Mon, 6 Dec 2010, [ISO-8859-1] Carlos Henrique Coêlho de Andrade wrote:
Dear Sven and Allin,
Thanks for your attention. The problem here is that all data rows have
missing values.
I think it would be better if I could send my data set to the list.
You're welcome to do so -- or if the file is too big for the list,
you could send it off-list. However, if all rows have missing
values it's hard to see how one could do any useful statistical
calculations using the data in question.
Allin Cottrell
4:26 a.m.
New subject: Building Matrix Using the Dataset
Em 6 de dezembro de 2010 Allin Cottrell <cottrell(a)wfu.edu> escreveu:
On Mon, 6 Dec 2010, [ISO-8859-1] Carlos Henrique Coêlho de Andrade wrote:
> Dear Sven and Allin,
>
> Thanks for your attention. The problem here is that all data rows have
> missing values.
>
> I think it would be better if I could send my data set to the list.
You're welcome to do so -- or if the file is too big for the list,
you could send it off-list. However, if all rows have missing
values it's hard to see how one could do any useful statistical
calculations using the data in question.
I'd attached my current data set. If I could fill all the missing values
with
some number (e.g. 9999) I would solve my problem using the command:
matrix A = { dataset }
Is there a way to fill the missing observations with some arbitrary number
using Gretl scripting?
Many thanks,
--
Henrique C. de Andrade
Doutorando em Economia Aplicada
Universidade Federal do Rio Grande do Sul
www.ufrgs.br/ppge
8:52 a.m.
New subject: Building Matrix Using the Dataset
Am 07.12.2010 05:26, schrieb Henrique Andrade:
I'd attached my current data set. If I could fill all the missing values
with
some number (e.g. 9999) I would solve my problem using the command:
matrix A = { dataset }
Is there a way to fill the missing observations with some arbitrary number
using Gretl scripting?
maybe something like:
mynumber = 9999
matrix newA = misszero(A) + missing(A)*mynumber
Actually there may be a case to replace the misszero() function with
something slightly more general, e.g. missrepl() or frommiss() which
takes a second optional argument to determine the value with which the
missings should be replaced (and in analogy generalize zeromiss() to
setmiss() or tomiss() or whatever).
cheers,
sven
12:53 p.m.
New subject: Building Matrix Using the Dataset
Em 7 de dezembro de 2010 Sven Schreiber <svetosch(a)gmx.net> escreveu:
Am 07.12.2010 05:26, schrieb Henrique Andrade:
>
> I'd attached my current data set. If I could fill all the missing values
> with
> some number (e.g. 9999) I would solve my problem using the command:
>
> matrix A = { dataset }
>
> Is there a way to fill the missing observations with some arbitrary
number
> using Gretl scripting?
>
maybe something like:
mynumber = 9999
matrix newA = misszero(A) + missing(A)*mynumber
Dear Sven, I'm trying to follow your suggestion but it doesn't work. Please
take a look at my script:
<script>
*matrix* A = { *dataset *}
*scalar *mynumber = 9999
*matrix *newA = misszero(A) + missing(A)*mynumber
</script>
Best regards,
--
Henrique C. de Andrade
Doutorando em Economia Aplicada
Universidade Federal do Rio Grande do Sul
www.ufrgs.br/ppge
1:16 p.m.
New subject: Building Matrix Using the Dataset
Em 7 de dezembro de 2010 Henrique Andrade <henrique.coelho(a)gmail.com>escreveu:
Em 7 de dezembro de 2010 Sven Schreiber <svetosch(a)gmx.net> escreveu:
Am 07.12.2010 05:26, schrieb Henrique Andrade:
>
> >
> > I'd attached my current data set. If I could fill all the missing values
> > with
> > some number (e.g. 9999) I would solve my problem using the command:
> >
> > matrix A = { dataset }
> >
> > Is there a way to fill the missing observations with some arbitrary
> number
> > using Gretl scripting?
> >
>
> maybe something like:
>
> mynumber = 9999
> matrix newA = misszero(A) + missing(A)*mynumber
>
Dear Sven, I'm trying to follow your suggestion but it doesn't work. Please
take a look at my script:
<script>
*matrix* A = { *dataset *}
*scalar *mynumber = 9999
*matrix *newA = misszero(A) + missing(A)*mynumber
</script>
Dear Sven and Allin, I'd found a solution for my problem. Please take a look
at the
following script:
*<script>*
*scalar *mynumber = 9999
*list *variables = 1
*loop *i=2..144
variables += $i
*endloop*
*loop foreach *i variables
*series *$i = misszero($i) + missing($i)*mynumber
*endloop*
*matrix *A = { *dataset *}
*</script>*
Thanks for your help!
Best regards,
--
Henrique C. de Andrade
Doutorando em Economia Aplicada
Universidade Federal do Rio Grande do Sul
www.ufrgs.br/ppge
9:25 a.m.
New subject: Building Matrix Using the Dataset
El 07/12/10 14:16, Henrique Andrade escribió:
Dear Sven and Allin, I'd found a solution for my problem. Please take a
look at the
following script:
*<script>*
*scalar *mynumber = 9999
*list *variables = 1
*loop *i=2..144
variables += $i
*endloop*
*loop foreach *i variables
*series *$i = misszero($i) + missing($i)*mynumber
*endloop*
*matrix *A = { *dataset *}
*</script>*
Thanks for your help!
Best regards,
--
Henrique C. de Andrade
Dear Henrique,
another alternative for your script:
<script>
open Dados.gdt
scalar mynumber = 9999
list variables = dataset
loop foreach i variables
series $i = ok($i)*$i + !ok($i)*mynumber
endloop
matrix A = { variables }
</script>
--
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.ea3.ehu.es
5127
days inactive
5129
days old
5 comments
4 participants
participants (4)
-
Allin Cottrell -
Henrique Andrade -
Ignacio Diaz-Emparanza -
Sven Schreiber