Date: Mon, 16 Jun 1997 22:09:13 -0400
Reply-To: Ellen Hertz <eshertz@ACCESS.DIGEX.NET>
Sender: "SAS(r) Discussion" <SAS-L@UGA.CC.UGA.EDU>
From: Ellen Hertz <eshertz@ACCESS.DIGEX.NET>
Organization: DIGEX, Inc.
Subject: Re: How could I force to a procedure to take datas in groups from
one data set
Content-Type: text/plain; charset=us-ascii
Francisco Rodriguez wrote:
>
> Hi everybody,
>
> I want to process a data set to produce an output trough the procedure
> NLIN. I need to fit such data to a certain function in groups of 50
> elements, and store the parameters resulting in other data set.
>
> The problem is the following: how could I force a procedure to take the
> data in groups of 50 elements from just one data set, to process it, and
> repeat the same operation until there is not any data left in the data
> set.
>
> All the groups contains 50 elements, but also there are a couple of
> variables in the identifies each group. Both together identifies
> completed each group within the data set, maybe is easier to use this
> two identifier rather that the number of observation to select the
> groups.
>
> Any suggestion will be appreciated
> Thanks in advance,
> -
> _____________________________________________________________
> Francisco Rodriguez-Fernandez Phytopathology Division
> Doctoral Stundent on Biophysics Lourizan Forestry Research Centre
>
> Ph. +34.86.856400 Xunta de Galicia
> Fax. +34.86.856420 Apto.127 Pontevedra
> Francisco.Rodriguez@sfp.cifl.cesga.es Pontevedra C.P.36080
> http://www.sfp.cifl.cesga.es/ Galicia / Spain / European Union
If, by variables identifying each group, you mean that
there is a set of variables v1,..,vn (n >= 1) such that
all n agree for a pair of observations if and only if
they are in the same group, you can get exactly what you
want with a BY statement. In fact if you don't have that, it might
be the easiest approach to introduce a variable, say
INDVAR = INT((_N_-1)/50), for the purpose.
Regards,
Ellen Hertz
| 
