Impute missing values by Replicating Previous Value - sas

Question

My data is in this format:
Var1 Var1 Var1 Value1 Imputer_Value1 Value2 Imputer_Value2
A A1 A11 6 6 15 15
A A1 A11 9 9 14 14
A A1 A12 1 1 19 19
A A2 A12 1 16 16
A A2 A13 10 10 13 13
A A2 A13 4 4 . 13
B B1 B11 8 8 13 13
B B1 B11 9 9 17 17
B B1 B12 5 5 18 18
B B2 B12 . 5 12 12
B B2 B13 2 2 20 20
B B2 B13 1 1 . 20
I want to impute the missing values by replicating the previous value from the same above group. Can anyone please tell me how to do it? I tried to follow this option. But this does not have multiple value computation option.
data imputedData;
set mydata;
n=_n_;
if missing(Value1) then
do;
do until (not missing(value1));
n=n-1;
set mydata(keep=Value1) point=n; *second SET statement;
end;
end;
run;
Thanks!

Answer 1

If I'm understanding your question correctly, there are a few simple ways you can do this. The easiest being use of the lag and coalesce functions. Unfortunately the lag function can sometimes yield unexpected results when missing values are involved.
Here is an example using lag.
data want;
set have;
* The coalesce function returns the first non-missing value
* and the lag function returns the last value;
value1 = coalesce(value1, lag(value1));
value2 = coalesce(value2, lag(value2));
run;
If that does not work then you may have to use a retain statement.
data want;
set have;
retain val1 val2;
* If it's not the first record do;
if _n_ > 1 then do;
value1 = coalesce(value1, val1);
value2 = coalesce(value2, val2);
end;
val1 = value1;
val2 = value2;
run;