adapting moving average code for additional dimension

[csetzkorn]

Looking at the code taken from here to calculate the moving average for one dimension/category patient:

data ds1;
  do patient='A','B','C';
    do month=1 to 7;
      num=int(ranuni(0)*10);
      output;
      end;
    end;
run;

proc sort;
  by patient;
run;

%let n = 4;

data ds2;
  set ds1;
  by patient;
  retain num_sum 0;
  if first.patient then do;
    count=0;
    num_sum=0;
  end;
  count+1;
  last&n=lag&n(num);
  if count gt &n then num_sum=sum(num_sum,num,-last&n);
  else num_sum=sum(num_sum,num);
  if count ge &n then mov_aver=num_sum/&n;
  else mov_aver=.;
run;

title 'Moving average within BY-Group';
proc print;
run;

I am wondering if this could be adapted for an additional dimenension/category Cat1. A starting point would be:

data ds1;
  do patient='A','B','C';
  	do Cat1='C1','C2';
	    do month=1 to 7;
	      num=int(ranuni(0)*10);
	      output;
	    end;
	end;
  end;
run;

proc sort;
  by patient Cat1;
run;

Any help would be very much appreciated. Thanks a lot.

[Reeza]

1. Change BY statement in SORT (done & shown) and Data Step (DS2)

2. Any time you see a reference to first.patient change it to first.cat1

I think that's all...but untested.

---

@csetzkorn wrote:

Looking at the code taken from here to calculate the moving average for one dimension/category patient:

data ds1;
  do patient='A','B','C';
    do month=1 to 7;
      num=int(ranuni(0)*10);
      output;
      end;
    end;
run;

proc sort;
  by patient;
run;

%let n = 4;

data ds2;
  set ds1;
  by patient;
  retain num_sum 0;
  if first.patient then do;
    count=0;
    num_sum=0;
  end;
  count+1;
  last&n=lag&n(num);
  if count gt &n then num_sum=sum(num_sum,num,-last&n);
  else num_sum=sum(num_sum,num);
  if count ge &n then mov_aver=num_sum/&n;
  else mov_aver=.;
run;

title 'Moving average within BY-Group';
proc print;
run;

I am wondering if this could be adapted for an additional dimenension/category Cat1. A starting point would be:

data ds1;
  do patient='A','B','C';
  	do Cat1='C1','C2';
	    do month=1 to 7;
	      num=int(ranuni(0)*10);
	      output;
	    end;
	end;
  end;
run;

proc sort;
  by patient Cat1;
run;

Any help would be very much appreciated. Thanks a lot.

---

[csetzkorn]

Thanks. Are you sure the sort has only to be done for one dimension? Please note that the data generation code is just an example and the data can be in any order. Ultimately the MA is to be applied at the time dimension month.

[Reeza]

What do you mean the sort is by one dimension? Your proc sort showed two variables in the BY statement. That's what you need. In both BY statements, the data step and sort. 

And the fastest way is to answer this is to test it out 🙂