How To Seprate Character Variable and Numeric Variable

[SushilKumar]

Hi

I Created a Dataset In a Character Variable it contains character and Numeric Value,

I want to Split character and Numeric Value

I Have a dataset which is subject and result are character variable

subject	Result
1001	20
1002	>50
1003	Positive
1004	30.5

I Want the Dataset to be

subject	Result	Numeric	Character
1001	20	20
1002	>50		>50
1003	Positive		Positive
1004	30.5	30.5

Could Any one Help Me

[LinusH]

The easiest way (I think) is to input the value to the numeric column using best. format. If the result is missing, store the value in the character column.

You'll have some notes in the log about invalid arguments to the input function, but that is expected.

[RichardinOz]

Warning messages can be suppressed using ??

Numeric = input(result, ?? best.) ;

If numeric = . then character = result ;

Richard

[LinusH]

Forgot that:smileyblush:.

Works only if you are using the data step, not in SQL.

[RichardinOz]

In my recollection proc SQL the input function does not generate warning messages in this situation, so the ?? is not necessary anyway.  The OP can use SQL to achieve the same result but in this instance a data step is the simplest solution.

Richard

[SushilKumar]

Could You write the Code

[RichardinOz]

The code is in my reply.

You need to add a data statement, a set statement, and a run statement at the end.

Richard

[FabienLACREUSE]

Dear

I very simple way

data x;  length Result$ 20;
Result='20';output;
Result='>50';output;
Result='123Position';output;
Result='.';output;
Result='.23';output;

run;

data x; set x; v=(verify(Result,'1234567890.'))-1;l=length(trim(result));
length Numeric 8. CHARACTER $20;
if v=l then NUMERIC=result;
  else CHARACTER=result;
run;

You can also investigaute into Regular expression, if you have some experience.

[Haikuo]

Your method works if there is no other special scenarios, such as scientific notations, for example "12E4".  You maybe able to exhaust all of the possibilities, but the best practice is to let SAS to worry about it. I vote for Richard's solution.

Haikuo