Well, typically that is done by addressing the variance in your population.  If you have a population in which there is wide variation and in which you could get a large number of different responses, you need a larger sample size/more surveys. If your population is largely the same, a smaller sample size/fewer survey will get the job done statistically.  Yes, it is a "cart before the horse" thing.  They want you to know your population variance to choose your sample size but you don't know the variance without a sample of  the population.    In the real world, there are also political concerns.   A sample/number of surveys that is perfectly acceptable statistically in terms of producing data sufficient to characterize the population may not be acceptable politically to those receiving the data.  As an example, in a population of 10,000 with almost no variance 100 surveys may be enough to tell you everything you need to know very accurately, but those receiving the data may not be satisfied that only 100 out of 10,000 things were examined.   We don't really know what you are surveying/sampling or anything about your context so it is hard to give specific advice.  In general, sample sizes can be too small, but they can't really be too big (they are just more work).      ... View more

Are you essentially trying to choose a sample size? ... View more

Give us an input data set. ... View more

See Bruno's answer here for part of that.   https://communities.sas.com/t5/SAS-Programming/Finding-Location-of-SAS-Macros/td-p/292642     ... View more

I'm reasonably certain I don't understand your calculation method (months x rate but then rate divided by days- why???), but here is something to maybe get you thinking.     Using Excel I can do this:   d e f g h i j k l m subscriber start end amount now mdiff cam mddiff cdam tam 1 1/1/2021 3/14/2021 200 3/14/2021 2 400 14 14.29 414.29 1 3/15/2021   300 11/23/2021 8 2400 9 33.33 2433.33 2 1/1/2021 1/14/2021 400 1/14/2021 0 0 14 28.57 28.57 3 1/15/2021   500 11/23/2021 10 5000 11 45.45 5045.45   Now is the end date or the current date where the end date is empty.   = IF(F7, F7,NOW()) mdiff is the difference between either the start and end or start and now in full months. =IF(F7,DATEDIF(E7,F7,"m"),DATEDIF(E7,H7,"m")) cam is a calculated monthly amount of full months times amount. =+I7*G7 mddiff is the difference between either the start and end or start and now in days minus the full months.  =IF(F7,DATEDIF(E7,F7,"md")+1, DATEDIF(E7,H7,"md")+1) cdam is the calculated daily amount =+G7/K7 tam is total amount and sum of cam and cdam =+J7+L7   With data modified like that, I think the complete charge per subscriber is the sum of tam grouped by subscriber, which can be done with Proc Summary or Proc SQL or something.   Edit:  Apparently there are issues with MD.  Who knew? Looks fixable. https://support.microsoft.com/en-us/office/datedif-function-25dba1a4-2812-480b-84dd-8b32a451b35c   ... View more

This looks as much a database design problem as anything else.   I see subject name, pay code, stream code, start date and end date all both tables, which looks bad on the face of it from here.  A table structure with a student table (primary key student id, perhaps demographic elements) and an employment table (multiple records per student allowed, with start dates, end dates, stream, and pay code for each record) might make things easier.    Perhaps some sample data could be provided and the desired result shown based on the provided data?  ... View more

Not sure what you are asking- could you explain more? Are you asking about how to screen your data before you run gproject on it?   data have; input id lat long; datalines; 1 45.5 73.5 2 55.5 63.5 3 65.5 53.5 4 75.5 43.5 5 85.5 33.5 ; run; proc gproject latlon degrees westlong project=proj4 to="EPSG:32618" data=have out=want; id id; run; ... View more

Apparently I really don't understand what is being asked here.   data stuff; input ss zz a_abc $ a_qwerty $ a_xyz $ abcde; datalines; 12 13 able charlie echo 16 14 15 baker delta foxtrot 17 ; run; data stuff2; input ss zz a_abc $ a_qwerty $ a_xyz $ abcde; datalines; 12 13 rock just owl 18 14 15 cooperative modernize cassette 19 ; run; proc sql; select a.ss, a.zz, a.a_abc, a.a_qwerty, a.a_xyz, a.abcde, b.a_abc as b_abc, b.a_qwerty as b_qwerty, b.a_xyz as b_xyz from stuff a inner join stuff2 b on a.ss = b.ss and a.zz = b.zz; run; That code makes this output: ss zz a_abc a_qwerty a_xyz abcde b_abc b_qwerty b_xyz 12 13 able charlie echo 16 rock just owl 14 15 baker delta foxtrot 17 cooperat moderniz cassette   but I somehow sense that is not what is truly wanted.      ... View more