topic Re: Create new variable based off of previous values in SAS Programming

Create new variable based off of previous values

lalaktgrau — Thu, 18 Oct 2018 14:59:21 GMT

Hi,

I'd like to create a variable that =1 when:
moving starts (first instance of moving=1) and continues being 1 until disease has continued for 3 minutes.Then, it becomes zero again.
Therefore, it only captures the onset of moving and the first 3 minutes of disease.

data WORK.CLASS;
infile datalines truncover;
input Name:$8. moving disease datetime;
informat datetime datetime20.;
format datetime datetime20.;
datalines;
Alfred 0 0 04MAR14:23:55:00
Alfred 0 0 04MAR14:23:56:00
Alfred 1 0 04MAR14:23:57:00
Alfred 1 0 04MAR14:23:58:00
Alfred 1 0 04MAR14:23:59:00
Alfred 1 1 05MAR14:00:00:00
Alfred 1 1 05MAR14:00:01:00
Alfred 1 1 05MAR14:00:02:00
Alfred 1 1 05MAR14:00:03:00
Alfred 1 1 05MAR14:00:04:00
mary 0 0 04MAR14:23:55:00
mary 0 0 04MAR14:23:56:00
mary 0 0 04MAR14:23:57:00
mary 0 0 04MAR14:23:58:00
mary 0 0 04MAR14:23:59:00
mary 1 0 05MAR14:00:00:00
mary 1 1 05MAR14:00:01:00
mary 1 1 05MAR14:00:02:00
mary 1 1 05MAR14:00:03:00
mary 1 1 05MAR14:00:04:00;

So if the variable was created properly, it would look like this:

Alfred 0 0 04MAR14:23:55:00 0 
Alfred 0 0 04MAR14:23:56:00 0 
Alfred 1 0 04MAR14:23:57:00 1 
Alfred 1 0 04MAR14:23:58:00 1 
Alfred 1 0 04MAR14:23:59:00 1 
Alfred 1 1 04MAR14:00:00:00 1 
Alfred 1 1 04MAR14:01:01:00 1 
Alfred 1 1 04MAR14:01:02:00 1 
Alfred 1 1 04MAR14:01:03:00 0 
Alfred 1 1 04MAR14:01:04:00 0

Re: Create new variable based off of previous values

ChrisNZ — Wed, 17 Oct 2018 05:04:16 GMT

1.Why is everything 04MAR14 even after midnight?

2.Why is Mary excluded when it matches (first instance of moving=1) and continues being 1 until disease has continued for 3 minutes

Re: Create new variable based off of previous values

lalaktgrau — Wed, 17 Oct 2018 13:58:09 GMT

It was just an example of times. This isn't my actual data, though it is very similar.

I just gave Alfred as a sample.

Re: Create new variable based off of previous values

mkeintz — Wed, 17 Oct 2018 14:39:10 GMT

The sample may be synthetic, which is fine, but it doesn't even follow your description. Specifically, it is not in chronological order. This does not help us help you.

Re: Create new variable based off of previous values

lalaktgrau — Wed, 17 Oct 2018 14:44:32 GMT

oh! Sorry. I had changed the datepart for the creation of another variable. I edited it! It should be good now!

Re: Create new variable based off of previous values

mkeintz — Wed, 17 Oct 2018 15:21:38 GMT

Now your data series skips from midnight to 1 hour after midnight. Is that intentional? I've changed them from 01:01:00, 01:02:00 to 00:01:00, 00:02:00, etc:

data WORK.CLASS;
infile datalines truncover;
input Name:$8. moving disease datetime;
informat datetime datetime20.;
format datetime datetime20.;
datalines;
Alfred 0 0 04MAR14:23:55:00
Alfred 0 0 04MAR14:23:56:00
Alfred 1 0 04MAR14:23:57:00
Alfred 1 0 04MAR14:23:58:00
Alfred 1 0 04MAR14:23:59:00
Alfred 1 1 05MAR14:00:00:00
Alfred 1 1 05MAR14:00:01:00
Alfred 1 1 05MAR14:00:02:00
Alfred 1 1 05MAR14:00:03:00
Alfred 1 1 05MAR14:00:04:00
mary 0 0 04MAR14:23:55:00
mary 0 0 04MAR14:23:56:00
mary 0 0 04MAR14:23:57:00
mary 0 0 04MAR14:23:58:00
mary 0 0 04MAR14:23:59:00
mary 1 0 05MAR14:00:00:00
mary 1 1 05MAR14:00:01:00
mary 1 1 05MAR14:00:02:00
mary 1 1 05MAR14:00:03:00
mary 1 1 05MAR14:00:04:00
run;


data want (drop=dis_dt);
  set class;
  by name;
  format dis_dt datetime20.;
  retain dummy dis_dt;
  if first.name then do;
     dummy=0;
     dis_dt=constant('bigint');  /* Make a dummy date way in the future*/
  end;
  if moving=1 then dummy=1;
  if dif(disease)=1 then dis_dt=datetime;
  if datetime>dis_dt+'00:03:00't then dummy=0; /*Set actual disease date when encountered*/
run;
proc print;run;

This program assumes:

moving always starts no later than disease
Only moving start or disease starts they never go back to zero.

Re: Create new variable based off of previous values

lalaktgrau — Thu, 18 Oct 2018 15:00:42 GMT

This is great, thank you!

Quick question: Is there a way to include a DO UNTIL loop that would stop processing (that day for that person) after we reach the first consecutive 3 minutes of disease?

In other words, if later that night disease becomes 0 and then goes back to 1--I don't want to count them.

data WORK.CLASS;
infile datalines truncover;
input Name:$8. moving disease datetime;
informat datetime datetime20.;
format datetime datetime20.;
datalines;
Alfred 0 0 04MAR14:23:55:00
Alfred 0 0 04MAR14:23:56:00
Alfred 1 0 04MAR14:23:57:00
Alfred 1 0 04MAR14:23:58:00
Alfred 1 0 04MAR14:23:59:00
Alfred 1 1 05MAR14:00:00:00
Alfred 1 1 05MAR14:00:01:00
Alfred 1 1 05MAR14:00:02:00
Alfred 1 0 05MAR14:00:03:00
Alfred 1 1 05MAR14:00:04:00
mary 0 0 04MAR14:23:55:00
mary 0 0 04MAR14:23:56:00
mary 0 0 04MAR14:23:57:00
mary 0 0 04MAR14:23:58:00
mary 0 0 04MAR14:23:59:00
mary 1 0 05MAR14:00:00:00
mary 1 1 05MAR14:00:01:00
mary 1 1 05MAR14:00:02:00
mary 1 1 05MAR14:00:03:00
mary 1 1 05MAR14:00:04:00

run;