Date: Mon, 1 Aug 2011 02:16:19 -0400
Reply-To: Søren Lassen <s.lassen@POST.TELE.DK>
Sender: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From: Søren Lassen <s.lassen@POST.TELE.DK>
Subject: Re: Getting output from one data set using vars listed in another
Content-Type: text/plain; charset=ISO-8859-1
Eduardo,
The simplest solution is probably to use SQL to generate an SQL statement,
e.g.:
/* Sample variable name list */
data vars;
input varname $;
cards;
N001
N002
N023
;run;
/* sample data */
data data;
array N0(*) 8 N001 N002 N023;
do j=1 to 200;
do i=1 to dim(n0);
n0(i)=ceil(ranuni(0)*5);
end;
output;
end;
keep N0:;
run;
proc sql noprint;
/* Create statement to create statistics table */
select catx(' ','select distinct',
quote(trim(varname)),'as name,',
varname,'as value,'
'count(*) as N from data group by',varname)
into:stmt separated by ' union ' from vars;
/* create statistics table */
create table want as &stmt;
quit;
Regards,
Søren
On Fri, 22 Jul 2011 16:23:52 -0400, Eduardo Galvan <egalvan@ISR.UMICH.EDU>
wrote:
>The data in rangevars will simply have a variable with the field name
(vname) and will have values such as these:
>N001
>N002
>N023.
>
>I want to run frequencies on the variables (N001, N002, N023, etc.) in the
second file (Section_N) so that I can have a list of all the values for
each variable in the Section_N file. I am going to use this information to
determine the min and max value for each variable. I'd run a proc means on
each, but I need to merge other data (value labels) to the table containing
the frequency output because I need to eventually exclude some values--
these vary across the different variables) when determining the min and max
values for each variable.
>
>So I want my output to look like:
>
>Variable Name Value Frequency
>N001 1 15
>N001 2 30
>N001 3 20
>N002 1 20
>N002 5 30
>N002 9 15
>N023 1000 10
>N023 1500 11
>N023 9993 21
>
>Your help is appreciated.
>
>Thanks.
>
>
>
>
>-----Original Message-----
>From: Data _null_; [mailto:iebupdte@gmail.com]
>Sent: Friday, July 22, 2011 4:10 PM
>To: Eduardo Galvan
>Cc: SAS-L@listserv.uga.edu
>Subject: Re: Getting output from one data set using vars listed in another
>
>You don't need no stinkin' EXCEL to write SAS programs. :-) SAS has all
kinds of code gen facilities but you may not need them.
>
>I will need example of the data you describe. I think I can picture it ok
but I am too lazy to try and figure out exactly what your data looks like
and the mock it up. Also show example of the output you are looking for.
>
>I'm not sure you will need to create individual files of each variable
>summary. SAS will create a nice file with all variables. Better to
>keep it all together.
>
>
>On Fri, Jul 22, 2011 at 2:53 PM, Eduardo Galvan <egalvan@isr.umich.edu>
wrote:
>> Hi,
>>
>> I have a data file (rangevars) that contains a list of certain
>> variables that appear in a second data file (Section_N). I want to
>> run frequencies on each variable and output the results for each to an
>> individual file using the variable name as the file name before I
>> merge all the individual files into a "master" file.
>>
>> Since I am not very advanced in my programming skills, I thought that
>> I would run a proc print on the rangevars table to get the list of
variables.
>> At that point I would copy the output to Excel and manipulate it to
>> create macro code to do what I want. However, I would like to learn
>> how to have this be entirely automated.
>>
>> Thank you.
>>
| 
