Date: Wed, 4 Oct 2006 02:01:33 -0700
Reply-To: "Nordlund, Dan (DSHS)" <NordlDJ@DSHS.WA.GOV>
Sender: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From: "Nordlund, Dan (DSHS)" <NordlDJ@DSHS.WA.GOV>
Subject: Re: Surprising retain feature
In-Reply-To: <1159945834.124060.182280@k70g2000cwa.googlegroups.com>
Content-Type: text/plain; charset=utf-8
> -----Original Message-----
> From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of avi
> Sent: Wednesday, October 04, 2006 12:11 AM
> To: SAS-L@LISTSERV.UGA.EDU
> Subject: Re: Surprising retain feature
>
> Hello,
>
> The problem i had was w. large datasets. I tried to reconstitute it
> with a small exemple, but here the retain feature does not work. I'm
> quite disturbed (in the exemple, the DPT value for resp_id=2 in the ALL
> dataset is supposed to retain its value for previous observation, but
> it did not...)
>
>
> ods html file="eee .html" path="C:\" nogtitle style=styles.statdoc ;
>
> data one;
> input resp_id q316 ;
> cards;
> 1 1
> 2 0
> 3 5
> ;
> proc print;
> data two;
> input resp_id ;
Was this intended to only read one variable here?
> cards;
> 4 1
> ;
> data all;
> attrib dpt format=$5.;
> set two one;
> if Q316=1 then dpt='1111 ';
> if Q316=2 then dpt='2222';
> if Q316=3 then dpt='3333';
> proc print;
> run;
> ods html close;
>
>
> Avi
Avi,
The variable, DPT, in your example is not retained either explicitly or implicitly. It is not listed explicitly in a RETAIN statement. It does not exist in either of the datasets listed in the SET statement and therefore it is not retained implicitly. So, the output is exactly as I would expect. DPT is set to missing at the beginning of each iteration through the data step.
I am too tired to think straight right now, but if someone doesn't beat me to it, I will try and give a short example illustrating implicit RETAINing in the morning after I get some sleep.
Dan
Daniel J. Nordlund
Research and Data Analysis
Washington State Department of Social and Health Services
Olympia, WA 98504-5204
| 
