cumulative frequency by group(when there is a missing value)

[sahoositaram555]

Hi,  i have a data set like below. fluidvalue is a numeric column having values xx, 0 and .(missing values) .for some patid's its not present. i tried doing it by retain statement. it works well if there is no .(missing value), but its stops counting especially by not considering the values after . (missing numeric dot).  

data a;

input patid fluidvalue;

cards;

01001 30

01001 34

01001 0

01001 23

01001 .

01001 .

01001 .

01001 .

01001 80

01001 0

01002 .

01002 23

01002 34

01002 .

01002 45

01002 0

01002 .

01002 89

01002 0

01002 .

01003 23

01003 45

01003 45

01003 56

01003 90

01004 0

01004 90

01004 98

01004 23

01004 0

01005 0

01005 0

01005 0

01005 23

;

run;

data result;set a;

retain cum_EXDOSE;
if first.patid then cum_EXDOSE=fluidvalue;
else cum_EXDOSE=cum_EXDOSE+fluidvalue;
by patid;

run;

Please suggest how to do modification to this program to get the final output.I would like to solve it by using this retain statement, if possible . it would be better if someone suggests me if there is a way to get the final value next to patid across all the observations(if 01003 has 7 observation and the cum_EXDOSE is 120 then for every observation of 01003 , cum_EXDOSE  will be 120 for all the rows of 01003 ).

[sahoositaram555]

Hi,  i have a data set like below. fluidvalue is a numeric column having values xx, 0 and .(missing values) .for some patid's its not present. i tried doing it by retain statement. it works well if there is no .(missing value), but its stops counting especially by not considering the values after . (missing numeric dot).  

data a;

input patid fluidvalue;

cards;

01001 30

01001 34

01001 0

01001 23

01001 .

01001 .

01001 .

01001 .

01001 80

01001 0

01002 .

01002 23

01002 34

01002 .

01002 45

01002 0

01002 .

01002 89

01002 0

01002 .

01003 23

01003 45

01003 45

01003 56

01003 90

01004 0

01004 90

01004 98

01004 23

01004 0

01005 0

01005 0

01005 0

01005 23

;

run;

data result;set a;

retain cum_EXDOSE;
if first.patid then cum_EXDOSE=fluidvalue;
else cum_EXDOSE=cum_EXDOSE+fluidvalue;
by patid;

run;

Please suggest how to do modification to this program to get the final output.I would like to solve it by using this retain statement, if possible . it would be better if someone suggests me if there is a way to get the final value next to patid across all the observations(if 01003 has 7 observation and the cum_EXDOSE is 120 then for every observation of 01003 , cum_EXDOSE  will be 120 for all the rows of 01003 ).

[PeterClemmensen]

Please do not double post. See my answer to your original post here

https://communities.sas.com/t5/SAS-Programming/cumulative-frequency-by-group-when-there-is-a-missing...

[PeterClemmensen · Accepted solution]

I'm not entirely sure here, but I think this is what you want

data want;
   do _N_ = 1 by 1 until (last.patid);
      set a;
      by patid;
      sum_fluidvalue = sum(sum_fluidvalue, fluidvalue);
   end;
   do _N_ = 1 to _N_;
      set a;
      output;
   end;
run;