Replace missing in a by group context

[blackandwhite]

Hi,

I have attached here the code to the dataset that I have and what I am trying to get to but have some difficulty. Essentially, I have a dataset with multiple visits (visit_no) by ID (different IDs have different number of total visits). For each ID, I am trying to fill in the missing values for "result" as "positive" for all visits following a visit in which this variable was already recorded as being "positive". Note: only looking forward when replacing the missing as positive. Note2: if result is negative, leave missing as is.  

Many thanks in advance for your help

**

data have;

input ID visit_no result $;

datalines;

1 1 positive

1 2 .

1 3 .

1 4 positive

2 1 negative

2 2 negative

2 3 positive

2 4 .

2 5 .

3 1 .

3 2 negative

3 3 .

4 1 .

4 2 positive

;

run;

data want;

input ID visit_no result $;

datalines;

1 1 positive

1 2 positive

1 3 positive

1 4 positive

2 1 negative

2 2 negative

2 3 positive

2 4 positive

2 5 positive

3 1 .

3 2 negative

3 3 .

4 1 .

4 2 positive

;

run;

[novinosrin · Accepted solution]

data have;

input ID visit_no result $;

datalines;

1 1 positive

1 2 .

1 3 .

1 4 positive

2 1 negative

2 2 negative

2 3 positive

2 4 .

2 5 .

3 1 .

3 2 negative

3 3 .

4 1 .

4 2 positive

;

run;

data want;
 set have;
 by id;
 retain t;
 length t $20;
 if first.id then t=' ';
 if not missing(result) then t=result;
 else if t='positive' then result=t;
 drop t;
run;

[blackandwhite]

thanks so much @novinosrin