topic Re: Hold value down to next missing rows in New SAS User

Hold value down to next missing rows

ashkum — Wed, 03 Oct 2018 09:58:58 GMT

id name salary

1 ash 1500

2 ani *

3 ayu 2000

4 shw *

5 ami 1200

6 su *

in the above question I want to replace * with just previous value. what would be the syntax in sas?

e.g In 2nd observation * is replaced by 1500 and in 4th observation * is replaced by 2000 and so on.

Re: Hold value down to next missing rows

RW9 — Wed, 03 Oct 2018 10:00:45 GMT

Something like:

data want (drop=lst);
  set have;
  retain lst $200;
  lst=ifc(salary ne "*",salary,lst);
  salary=ifc(salary="*",lst,salary);
run;

Note not tested, I assume salary is character as * is not numeric, post test data in the form of a datastep in future.

Re: Hold value down to next missing rows

ashkum — Wed, 03 Oct 2018 10:22:41 GMT

salary is nuemric here

Re: Hold value down to next missing rows

RW9 — Wed, 03 Oct 2018 10:27:28 GMT

Numeric variables cannot hold the symbol *, therefore either the variable is character or you have changed the missing option. Either way, this is why we always ask for test data in the form of a datastep to avoid these needless backwards and forwards guessing your data.

Re: Hold value down to next missing rows

FreelanceReinh — Wed, 03 Oct 2018 12:26:52 GMT

@ashkum wrote:

salary is nuemric here

Then just edit RW9's code accordingly: Delete (and forget) $200, use ifn in place of ifc and replace "*" by . (the period denotes a numeric missing value) or whatever numeric value was displayed as an asterisk (assuming it was always the same value).