topic Re: SAS - remove empty rows only from one variable without changing the other variables in the datas in New SAS User

SAS - remove empty rows only from one variable without changing the other variables in the dataset

adrfinance — Mon, 27 Jan 2020 11:00:40 GMT

I have a variable that looks like:

12341234

12341234123

123412341234

12314123124

I want to keep only the rows with values, i.e. remove the rows without values. How do I do that only for this variable, without changing the other variables in the dataset?

Re: SAS - remove empty rows only from one variable without changing the other variables in the datas

Amir — Mon, 27 Jan 2020 11:06:06 GMT

Hi @adrfinance,

Is the variable type numeric or character?

And are you happy to lose the data for the other variables in the same rows where this variable has no value?

Regards,

Amir.

Re: SAS - remove empty rows only from one variable without changing the other variables in the datas

Amir — Mon, 27 Jan 2020 11:17:17 GMT

If it is numeric and you are happy to lose the values of the other variables where your variable is missing then you could try the following:

data have;
   infile datalines dsd;
   input
      seq_id : $8.
      my_var :  8.
   ;

   datalines;
1,12341234
2,.
3,.
4,.
5,12341234123
6,.
7,.
8,.
9,123412341234
10,.
11,.
12,.
13,12314123124
;


data want;
   set have;
   where not missing(my_var);
run;

If you need something else then please please provide what output you require.

Thanks & kind regards,

Amir.

Re: SAS - remove empty rows only from one variable without changing the other variables in the datas

adrfinance — Mon, 27 Jan 2020 12:12:21 GMT

the variable is numeric and I would like to keep the other variables as they are

Re: SAS - remove empty rows only from one variable without changing the other variables in the datas

Amir — Mon, 27 Jan 2020 12:22:28 GMT

Thanks for the confirmation.

Does the code I posted give you what you want?

If not then please provide example output data for the given input data.

Kind regards,

Amir.