the chance that anyone of the 3 being added to your sample with a reply of yes is 0. 1/15 would possibly answer yes.   ... View more

proc sql; select count(distinct player)   from have   group by team,month; quit; ... View more

You have 4 groups, you can use either ANOVA or a regression model. ... View more

@emma19901   Help us help you.  Please show the layout of the desired result data. ... View more

When posting code and/or log always paste the text into the window opened by clicking on the {i}-icon. This way formatting is preserved, increasing readability. ... View more

No idea what a CCI index is but I would guess the following code could be close:   data risk; set datasets.riskdata; if diagnosis= 'TYPE1_DM' then diagnosis='y'; if diagnosis='TYPE2_DM' THEN DIAGNOSIS='n'; if diagnosis= 'PXECTMY' then diagnosis='n'; if los<0 then los=.; array A{10} diagnosis dialysis lipiddis cbv_ds pvd neuropthy respprob cancrpre cbypass mi; do i=1 to 10; A{i} = upcase(A{i}); if A{i} not in ("Y", "N") then call missing(A{i}); end; array c{10} (2,3,2,1,1,1,1,3,1,2); CCI = 0; do i = 1 to 10; if A{i} = 'Y' then CCI = CCI + c{i}; end; drop i; run; ... View more

For your first question, you will need to define a new set of variables.  Character variables won't hold numeric values.  One possibility:   data hw2;  set datasets.riskdata;  array a {5} cbypass cbv_ds respprob neuropthy pvd;  array b {5} cbypass_num cbv_ds_num respprob_num neuropthy_num pvd_num;  do i=1 to 5;  if upcase(a{i})= "Y" then b{i}=1;  else b{i}=0;  end;  run;   Notice that when you omit the quotes, y refers to a variable name.  When you add quotes, "Y" refers to a character string.  This will impact your second question where you check whether a variable is equal to delete.  With no quotes, delete is the name of a variable.  Clearly, this is not the comparison you want to make.  However, your objective in the second DATA step is not clear.  What are you trying to accomplish?  Describe the results you want, don't show a program.  It's really not clear what you want when you say "erase missing values". ... View more

WHere's the rest of the code? The PROC POWER statement for starters?   73 proc groupweights=   And the rest of that code doesn't look close to correct.    See the documentation reference here: http://documentation.sas.com/?docsetId=statug&docsetTarget=statug_power_toc.htm&docsetVersion=14.3&locale=en   See the examples section for various full examples.    @emma19901 wrote: Hi, I am trying to calculate power for a hypothetical cohort study, I have run in multiple times this code but I still receive the same error 180-322.   Thanks in advance!   groupweights=(1 9)/*Divide ntotal 9 to 1 between the groups => n2=9*n1, where n1 = exposed sample size & n2 = unexposed sample size*/ ntotal=120000 /*120,000 total participants in the study*/ power=.; /*Calculate power*/   OPTIONS NONOTES NOSTIMER NOSOURCE NOSYNTAXCHECK; 72 73 proc groupweights=(1|9)/*Divide ntotal 9 to 1 between the groups => n2=9*n1, where n1 = exposed sample size & n2 = 73 ! unexposed sample size*/ ERROR: Procedure GROUPWEIGHTS not found. 74 ntotal=120000 /*120,000 total participants in the study*/ 75 power=.; /*Calculate power*/ 76 77 OPTIONS NONOTES NOSTIMER NOSOURCE NOSYNTAXCHECK;   ... View more