topic Re: Flag variable in SAS Programming

Flag variable

luvscandy27 — Tue, 04 Feb 2020 00:05:40 GMT

I have the following data below. I would like to choose the obs where
first_year eq 1. I would like my data set to look like the table below
could someone assist me with this?

data data1;
infile datalines delimiter = ',';
input customerid $ year regdate:mmddyy10. completedate:mmddyy10. first_ssn
last_ssn first_year last_year;
format regdate completedate mmddyy10.;
datalines;
100, 2019,04/30/2019, 01/24/2018, 1,0,1,0
100,2018,12/31/2018,12/16/2019, 0,0,0,0
100,2018,11/30/2018,12/16/2019, 0,0,0,1
100,2019, 08/31/2019,12/16/2019,0,0,1,0
;
run;

customerid	year	regdate	completedate	first_ssn	last_ssn	first_year	last_year
100	2019	4/30/2019	1/24/2018	1	0	1	0
100	2019	8/31/2019	12/16/2019	0	0	1	0

Re: Flag variable

ballardw — Tue, 04 Feb 2020 00:11:11 GMT

Generic form to create a subset of data based on an existing variable for a given value:

Data want;

set have;

where variable=1;

run;

You may not even need to create a new data set as you can use data set options such as

Proc print data=have (where=(variable=1));

run;

to select the records to use as needed.

Re: Flag variable

luvscandy27 — Tue, 04 Feb 2020 00:21:11 GMT

The dataset above was created using first. last. I was trying to remove the obs I don't need using first.last.
I tried the code below but it does not keep all the obs where first_year eq 1:

data want;
set datal;
by ssn first_year;
if first.first_year;
run;

Re: Flag variable

Patrick — Tue, 04 Feb 2020 00:47:37 GMT

@luvscandy27 wrote:

The dataset above was created using first. last. I was trying to remove the obs I don't need using first.last.
I tried the code below but it does not keep all the obs where first_year eq 1:

data want;
set datal;
by ssn first_year;
if first.first_year;
run;

The IF FIRST.<variable name> syntax returns the first row of a sorted set and not all rows with a specific value.

Try one of the options below for what you want.

data data1;
  infile datalines delimiter = ',';
  input customerid $ year regdate:mmddyy10. completedate:mmddyy10. first_ssn
    last_ssn first_year last_year;
  format regdate completedate mmddyy10.;
  datalines;
100, 2019,04/30/2019, 01/24/2018, 1,0,1,0
100,2018,12/31/2018,12/16/2019, 0,0,0,0
100,2018,11/30/2018,12/16/2019, 0,0,0,1
100,2019, 08/31/2019,12/16/2019,0,0,1,0
;
run;

proc print data=data1(where=(first_year=1));
run;

data want1;
  set data1(where=(first_year=1));
run;

data want2;
  set data1;
  where first_year=1;
run;

data want3;
  set data1;
  if first_year=1;
run;