topic Re: Set numeric missing value as 'NA' in SAS Data Management

Set numeric missing value as 'NA'

zhouxysherry — Mon, 28 Nov 2016 17:41:27 GMT

Hey there, as the title says, how can I make the numberic missing value (format: BEST12., informat: 12.) showed as 'NA' in below data set?

I tried to use put() statement to transform numeric variables to character first, but the transformation doesn't work.

Thanks!

Re: Set numeric missing value as 'NA'

Haris — Mon, 28 Nov 2016 20:31:37 GMT

There's not enough detail in your post to determine exactly what you are trying to do and where you want 'NA' to show up: data view, report, some sort of procedure output?

Try to apply a format:
PROC FORMAT; value NA .='NA'; run;

Re: Set numeric missing value as 'NA'

zhouxysherry — Mon, 28 Nov 2016 21:56:14 GMT

I would like 'NA' to show in data view.

I tried PROC FORMAT, but it doesn't work.

Re: Set numeric missing value as 'NA'

Haris — Mon, 28 Nov 2016 22:20:01 GMT

You need to assign the format to your variables too. This code works for me. You may also need to tell your data viewer what to display: different machines have different defaults and can either show raw or formatted values.

proc format; value NA .='NA'; run;

data test;
	a = .; b=1; c=1; output;
	a = 1; b=.; c=1; output;
	a = 1; b=1; c=.; output;
	format a b c NA.;
run;

Re: Set numeric missing value as 'NA'

ballardw — Tue, 29 Nov 2016 00:03:11 GMT

The NUMERIC variables of your array ZERO are still numeric.

zero [i] = 'NA' will fail as 'NA' isn't acceptable. And that is what you are doing with the PUT statement.

try

data test;

set output_de21_de42;

format _numeric_ trouble.;

run;

Re: Set numeric missing value as 'NA'

David_Luttrell — Fri, 02 Dec 2016 16:53:00 GMT

Change the variable(s) to a characters

Use a case statement to apply the NA values in the transformation otherwise that value.

It may be more appropriate to transpose the dataset before applying as this will mean less cols to change.