Date:         Mon, 18 Oct 2010 19:56:36 -0400
Reply-To:     Nat Wooding <nathani@VERIZON.NET>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Nat Wooding <nathani@VERIZON.NET>
Subject:      Re: reading NA as missing values without errors
In-Reply-To:  <AANLkTimfPQoNbGxG_+RfW5Y7cpMgMgrFyDX-FN03RrPD@mail.gmail.com>
Content-Type: text/plain; charset="US-ASCII"

Or, you can use the Captain Kirk approach and fix the input data:

 data UnitedNations;
         infile datalines dsd dlm=" ";
                 input @;
                 _infile_ = tranwrd( _infile_ , 'NA' , '.' );
         input test $ test2  test3 ;
         datalines;
id1 3 6
id2 4 7
id3 NA 3
id4 NA NA
id5 4 NA
;;;;
 run;

Thanks for the sample code, Joe

Nat Wooding
-----Original Message-----
From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of Joe
Matise
Sent: Monday, October 18, 2010 2:21 PM
To: SAS-L@LISTSERV.UGA.EDU
Subject: Re: reading NA as missing values without errors

There's always the error ignoring option.

input numfield ??;

ie:


 data UnitedNations;
         infile datalines dsd dlm=" ";
         input test $ test2 ?? test3 ??;
         datalines;
id1 3 6
id2 4 7
id3 NA 3
id4 NA NA
id5 4 NA
;;;;
 run;

(Your http didn't pass through to me properly so I had to fake it).

-Joe


On Mon, Oct 18, 2010 at 12:51 PM, Arthur Tabachneck
<art297@netscape.net>wrote:

> Michael,
>
> One alternative is to save the file, in R, as a dbf file and then
> import the dbf file into SAS.  An example is shown at:
> http://www.r-bloggers.com/example-7-10-get-data-from-r-into-sas/
>
> HTH,
> Art
> ---------
> On Oct 18, 12:34 pm, Michael Friendly <frien...@yorku.ca> wrote:
> > In R data files (.txt or .csv), missing is represented as 'NA'.  In SAS,
> > the missing statement allows only a set of single characters to
> > be read and interpreted as missing values, without generating errors.
> >
> > For the following input step, there are lots of missing values in the
> > numeric variables represented as NA.  The data step below works, i.e.,
> > the missing NA values are correctly translated to SAS missing (.), but
> > generates a great many errors.
> >
> > Perhaps there is some solution using PROC FORMAT and INVALUE formats,
> > but I haven't found anything that applies in this case.
> >
> > title 'Missing data: UN example';
> >
> > filename UN URL
> > "http://socserv.socsci.mcmaster.ca/jfox/Books/Applied-Regression-2E/da..
> .";
> > data UnitedNations;
> >         infile UN dsd dlm=" " firstobs=2;
> >         input country $ region $ tfr contraception educationMale
> > educationFemale lifeMale lifeFemale
> >         IMR GDP economicActivityMale economicActivityFemale
> illiteracyMale
> > illiteracyFemale;
> > run;
> >
> > thanks,
> > -Michael
>