topic Re: excludes the lowest value when calculating mean in SAS? in SAS Programming

excludes the lowest value when calculating mean in SAS?

Uzeus7869 — Fri, 06 Sep 2019 03:31:20 GMT

So, My professor asked me to create a SAS program to excludes the lowest midterm grades when computing the prefinal grades(mean of exams), I google it, and it's have nothing help there, I can't find appropriate code.

Thank you for help

Re: excludes the lowest value when calculating mean in SAS?

ChrisNZ — Fri, 06 Sep 2019 03:51:12 GMT

So, My professor asked me to create a SAS program to excludes the lowest midterm grades when computing the prefinal grades(mean of exams), I google it, and it's have nothing help there, I can't find appropriate code.

Surely your professor also gave you hints as to how to do this.

Re: excludes the lowest value when calculating mean in SAS?

PGStats — Fri, 06 Sep 2019 03:51:19 GMT

How would you calculate the mean without excluding the lowest values?

Re: excludes the lowest value when calculating mean in SAS?

yabwon — Thu, 12 Sep 2019 17:21:09 GMT

Hi,

for small size data arrays sounds reasonable

all the best

Bart

options fullstimer;
data have;                                                                                                               
do nums=1 to 5555555;                                                                                                       
  num=int(10*uniform(1234))+2; 
    if mod(nums,17) = 0 then num =.; 
  output;                                                                                                             
end;                                                                                                                   
run;     

data _null_;
  if 0 then set have nobs=NOBS;
       call symputX('NOBS', NOBS,'G');
  stop;
run;


data want;
  array _temp_[&nobs.] _temporary_;

  x = .;
  do until (EOF);
    set have end=EOF curobs=curobs;
    _temp_[curobs] = num;
    x = x<>(-num);
  end;
  
  x = -x;

  if x = . 
  then avg = .;
  else do curobs = lbound(_temp_) to hbound(_temp_);
    if _temp_[curobs] > x then 
      do;
        s + _temp_[curobs];
        c + 1;
      end;
  end;

  avg = divide(s,c);
  output;
run;