topic Re: Replacing numeric missing variable value with 'NA' in SAS Programming

Replacing numeric missing variable value with 'NA'

subrat1 — Fri, 18 Aug 2017 19:06:37 GMT

data yourdata;
   set yourdata;
   array change _numeric_;
        do over change;
            if change=. then change= NA;
        end;
 run ;

The above sas code does not work as i am trying to replace missing Numeric variable value with NA

Re: Replacing numeric missing variable value with 'NA'

novinosrin — Fri, 18 Aug 2017 19:13:39 GMT

I think you would be better off using a user defined format to replace .=NA

Re: Replacing numeric missing variable value with 'NA'

Reeza — Fri, 18 Aug 2017 19:17:30 GMT

Well, a column is numeric or character and NA is a character. So you're trying to put a character in a field that expects a number.

Do you need NA to export it to R?

How are you exporting, perhaps modify it there?

What you're trying to do overall will define how this should be approached.

Re: Replacing numeric missing variable value with 'NA'

subrat1 — Fri, 18 Aug 2017 19:19:01 GMT

Even after using user defined format .....The format of all numeric
variable changes....hence the data does not show correct value

Re: Replacing numeric missing variable value with 'NA'

Reeza — Fri, 18 Aug 2017 19:20:57 GMT

No, That's not how a user defined format works. Post your code if you're having issues.

@subrat1 wrote:
Even after using user defined format .....The format of all numeric
variable changes....hence the data does not show correct value

Re: Replacing numeric missing variable value with 'NA'

SuryaKiran — Fri, 18 Aug 2017 20:58:07 GMT

Hi,

User defined format solves your problem. See below code

PROC FORMAT ;
PICTURE Num    .="NA"
             OTHER="00.00";
			 run;

data test;
input  Number ;
format Number num. ;
cards;
3.3
6.8
10
.
run;

Re: Replacing numeric missing variable value with 'NA'

PGStats — Sun, 20 Aug 2017 04:41:47 GMT

Defining your own numeric format is the simplest way to replace the printed values:

data test;
do x = 1, ., .a, ._, 2;
    output;
    end;
run;

proc format;
value myMiss
. = "          NA"
other = [best12.];
run;

proc datasets library=work nolist;
modify test;
format _numeric_ myMiss.;
run;

proc print data=test; run;

Note that special missing values are not affected by the custom format as defined above.