topic Re: How to use drop and rename in the data statement at one time in SAS Procedures

How to use drop and rename in the data statement at one time

jcis7 — Wed, 22 Mar 2017 23:41:20 GMT

Hi,

I'm trying to drop variables and rename variables in a dataset.

I have var1 and var2 in the orignal dataset and am converting them into percentages using arrays. I'd like the variables in the output dataset to have the same names var1 and var2 so I'm trying to rename the new variables I created (pctvar1 pctvar2) to become var1 and var2 and also drop the old var1 and var2 variables that I don't need.

data round (drop=var1 var2)

rename= (pctvar1=var1 pctvar2=var2);

set first;

*create percentages;

array mult var1 var2;

array pct pctvar1 pctvar2;

do over mult

pct=mult*100;

end;

*--round to nearest tenth;

do over pct;

pct=round(pct, .1);

end;

run;

What am I doing wrong? thanks!

Re: How to use drop and rename in the data statement at one time

collinelliot — Wed, 22 Mar 2017 23:50:03 GMT

Ignore my stupid calculations, but it looks like you want to do the following, in which case it's just a matter of getting your parentheses correct:

data first;
    var1 = 1;
    var2 = 1;
run;

data round(drop = var1 var2 rename= (pctvar1=var1 pctvar2=var2));
    set first;
    pctVar1 = var1 / 10;
    pctVar2 = var2 / 10;
run;

Re: How to use drop and rename in the data statement at one time

Tom — Thu, 23 Mar 2017 01:14:10 GMT

A little reformatting of your DATA statement will probably make the error more obvious.

data 
  round (drop=var1 var2)
  rename= (pctvar1=var1 pctvar2=var2)
;

You have one output dataset name ROUND with the DROP= dataset option.

Then it looks like you are starting to also define another output dataset named RENAME, but you added an extra = sign after the data set name.

So fixing the nesting of your () so that you have just one output dataset you get this syntax:

data round 
 (drop=var1 var2
  rename= (pctvar1=var1 pctvar2=var2)
 )
;

It probably is easier to type if you just use the DROP and RENAME statements instead of adding dataset options to the output dataset.

data round ;
  set .... ;
  ....
  drop var1 var2;
  rename pctvar1=var1 pctvar2=var2;
run;

Re: How to use drop and rename in the data statement at one time

jcis7 — Thu, 23 Mar 2017 01:21:38 GMT

Thank you both! Super helpful!!

Re: How to use drop and rename in the data statement at one time

vidyapedii — Tue, 12 May 2020 12:40:52 GMT

data first;
var1 = 1;
var2 = 1;
var3= 3;
run;

data round (drop=var3 rename=(var1=pctvar1 var2=pctvar2));
set first ;
Var1 = var1 / 10;
Var2 = var2 / 10;
run;

you dont need drop the variable its been already renamed

Re: How to use drop and rename in the data statement at one time

_maldini_ — Fri, 19 Nov 2021 20:41:17 GMT

What is the order of operations here? Doesn't this syntax drop variables that were renamed?
What syntax would allow you to drop the original variables after renaming them? In other words, pctvar1 gets renamed to var1, pctvar2 gets renamed to var2 and then pctvar1 and pctvar1 get dropped.

Re: How to use drop and rename in the data statement at one time

Tom — Fri, 19 Nov 2021 22:42:17 GMT

Dataset options are processed in the alphabetical order:

Drop=/Keep= first

Rename= second

Where= last