If anyone can help me to achieve the following, would be of great help. In this, I have excluded the requirement for subjid 103 v1 and v2.   data have; input subjid visit$ aval; cards; 101 screen 20 101 base 25 101 v1 25 101 v2 26 102 screen 25 102 base 30 102 v1 25 102 v2 30 103 screen 23 103 base 33 ; run;   There are 3 subjid's 101, 102 and 103. For subjid 101 and 102, the row with visit as base is highlighted in blue color. Since, subjid 103 does not have v1 and v2, I have not highlighted the base row is not highlighted.   I want the above blue color highlighted rows inserted in between within each subjid group as follows:   data want; input subjid visit$ aval; cards; 101 screen 20 101 base   25 101 v1     25 101 base   25 101 v2     26 102 screen 25 102 base   30 102 v1     25 102 base   30 102 v2     30 103 screen 23 103 base   33 ; run; Then newly inserted rows are highlighted in Amber color. ... View more

For 103 v1 and v2, since it's just for testing purpose, I have used some example values for aval. ... View more

It's that I want to insert a row containing the base value for subjid before their visit 2 i.e., v2. I am just checking how to carry forward this row and insert it.    If visit 1 and visit 2 are missing for any subjid (in this example, they are missing for subjid 103), then create these two rows for that subjid (here aval can have any value as it is just for test purpose) and insert the row with base visit before the row of visit 2. ... View more

Here, is the data set that I have: data have; input subjid visit$ aval; cards; 101 screen 20 101 base 25 101 v1 25 101 v2 26 102 screen 25 102 base 30 102 v1 25 102 v2 30 103 screen 23 103 base 33 ; run;   There are 3 subjid's 101, 102 and 103. For each of these subjid group values, the row with visit as base is highlighted in blue color.   I want the above blue color highlighted rows inserted in between within each subjid group as follows and for subjid 103 since there are no rows for visits v1 and v2 and hence need to add these rows also:   data want; input subjid visit$ aval; cards; 101 screen 20 101 base   25 101 v1     25 101 base   25 101 v2     26 102 screen 25 102 base   30 102 v1     25 102 base   30 102 v2     30 103 screen 23 103 base   33 103 v1     20 103 base   33 103 v2     25 ; run; Then newly inserted rows are highlighted in Amber color.   Thanks ... View more

Thank you very much Tom, for a very detailed explanation. I will try all these cases. ... View more

Thank you. I will try this. ... View more

Thank you very much for the reply. I will try if I can use this example to BOCF (Baseline Observation Carry Forward).   ... View more

Yes, you are right that it is Baseline Observation Carry Forward.  Any example code explaining this concept, will be a great help.    Thanks ... View more

Actually, it is Baseline Observation Carry Forward. As found in Google, it is defined as "The baseline observation carry forward (BOCF) method, in which baseline observations are treated as the final responses from patients without any post-baseline observation or patients who drop out after the first post-baseline treatment, has been suggested for certain such studies by some regulatory agencies.".  I don't have example data, but want help with this, if anyone can help me with an example of BOCF.   Thanks ... View more

Sure and once again thank you very much. ... View more

By removing $, the following worked : imputed_date = put(intnx('mon',input(cats('01',date_monthpart,date_yearpart),date9.),0,'e'),date9.); If there are any better ways to do this, please advise. ... View more

You are absolutely right that the code that I have mentioned has given the following warning message:   WARNING: Character format specified for the result of a numeric expression.   Thank you very much for pointing out and providing the code. However, I want the sasdate to be character but in the code provided by you it's numeric. Can you please advise how to achieve this?  I have tried using put function but since sasdate is numeric, getting warning messages for this also.   Once, again thank you very much for your valuable guidance. ... View more

Thank you for the reply. The imputation rules are: If date part is not valid then replace it with last day of the month in the given date.  For example, ukfeb2022 has invalid date part 'uk' and month is feb and year 2022 is not a leap year I should get it as 28FEB2022. If both date part and month part are not valid then it should be 31DEC for example xyunk2019 has invalid date part xy and invalid month part unk and hence it should be 31DEC2019.   For the first part, I have used the following and it is working: imputed_date = put(intnx('mon',input(cats('01',date_monthpart,date_yearpart),date9.),0,'e'),$date9.);   ... View more