BookmarkSubscribeRSS Feed
sahotah
Calcite | Level 5

Hello!

I'm trying to make this bit of code more robust:

if 99 le days le 182 then  visit='6 month';
else if days le ceil(((9*30.5)+14)) then visit='9 month';
else if days le ceil(((12*30.5)+14)) then visit='12 month';
else if days le ceil(((15*30.5)+14)) then visit='15 month';
else if days le ceil(((18*30.5)+14)) then visit='18 month';
else if days le ceil(((21*30.5)+14)) then visit='21 month';
else if days le ceil(((24*30.5)+14)) then visit='24 month';
else if days le ceil(((27*30.5)+14)) then visit='27 month';
else if days le ceil(((30*30.5)+14)) then visit='30 month';
else if days le ceil(((33*30.5)+14)) then visit='33 month';
else if days le ceil(((36*30.5)+14)) then visit='36 month';
else if days le ceil(((39*30.5)+14)) then visit='39 month';
else if days le ceil(((42*30.5)+14)) then visit='42 month';
else if days le ceil(((45*30.5)+14)) then visit='45 month';

And have been able to somewhat make it more robust using this method:

do i=9 by 3 until(last.id);

  set origvisits;

  days = VISITDTC -trtstdt;

  label days='Number of days since original ib1001 start';

  if 99 le days le 182 then  visit='6 month';

  by id ;

  %let n=i;

  if days le ceil(((i*30.5)+14)) then visit=strip(&n||' month');

  else if days le ceil((((3+i)*30.5)+14)) then visit=strip(&n+3||' month');

  else if days le ceil((((6+i)*30.5)+14)) then visit=strip(&n+6||' month');

  else if days le ceil((((9+i)*30.5)+14)) then visit=strip(&n+9||' month');

  else if days le ceil((((12+i)*30.5)+14)) then visit=strip(&n+12||' month');

  output;

end;

But its still is not robust enough.

What I've constructed above is each observation is being read in with i=9 and increasing by 3 until last.id. And then i increments sequentially for each observation within that by group (id, in this case)


What I really want SAS to do is read one observation at a time and keep trying values of i (the index of the do loop) until the condition if days le ceil(((i*30.5)+14))  is true, so that I can assign a nominal visit date.


For example, suppose the condition  days le (9*30.5)+14 is not true, I want SAS to stay with that same observation and try (12*30.5)+14, if that condition is not true then try (15*30.5)+14 until the days le (i*30.5)+14 condition is true. Then move on to the next observation in the dataset, with i being reset back to 9.


I hope this makes sense, I know what I want SAS to do...but can't quite articulate it.


Any help or guidance would be appreciated

3 REPLIES 3
dkb
Quartz | Level 8 dkb
Quartz | Level 8

The first two groups (6 and 9 months) don't follow the general rule so they have to be coded separately.  The others follow a simple rule based on the number of days; this is one way of many to approach it:

if 99 le days le 182 then  visit=' 6 month';

else if days le ceil(((9*30.5)+14)) then visit=' 9 month';

else do;

   nmon = 3 * ceil((days-14.5) / (3 * 30.5));

   if nmon le 99 then visit = put(nmon, 2.)!! ' month';

  else put 'ERROR: Danger, Will Robinson! Numeric field overflow!';

end;

dkb
Quartz | Level 8 dkb
Quartz | Level 8

Thanks Reeza. I hadn't realised the OP was spraying.

sas-innovate-2024.png

Don't miss out on SAS Innovate - Register now for the FREE Livestream!

Can't make it to Vegas? No problem! Watch our general sessions LIVE or on-demand starting April 17th. Hear from SAS execs, best-selling author Adam Grant, Hot Ones host Sean Evans, top tech journalist Kara Swisher, AI expert Cassie Kozyrkov, and the mind-blowing dance crew iLuminate! Plus, get access to over 20 breakout sessions.

 

Register now!

What is Bayesian Analysis?

Learn the difference between classical and Bayesian statistical approaches and see a few PROC examples to perform Bayesian analysis in this video.

Find more tutorials on the SAS Users YouTube channel.

Click image to register for webinarClick image to register for webinar

Classroom Training Available!

Select SAS Training centers are offering in-person courses. View upcoming courses for:

View all other training opportunities.

Discussion stats
  • 3 replies
  • 878 views
  • 1 like
  • 3 in conversation