filling missing data per id from cells below.

[Roshani Patel]

Dear Stata Community,

I am working with long panel data. I have multiple observations per id.

In my original data set I have had to insert a baseline "observation" row for time 0. However I am struggling to fill "empty" cells with information for the particant id such as gender, longest followup.

I suppose what I am asking is - is there a way I can fill information for specific variables using the cell information from observation "2" for a specific variable and use it to fill in the currently missing observation "1".

e.g I would like to carry the gender information for the id into observation "1" which is currently missing.

I hope this question makes sense.

Many thanks again for all your help,

ROshani

----------------------- copy starting from the next line -----------------------

* Example generated by -dataex-. To install: ssc install dataex
clear
input float(obsnum observation) int id byte(gender first_op) float longest_fu
 1  1 1 . .  .
 2  2 1 1 1 26
 3  3 1 1 1 26
 4  4 1 1 1 26
 5  5 1 1 1 26
 6  6 1 1 1 26
 7  7 1 1 1 26
 8  8 1 1 1 26
 9  9 1 1 1 26
10 10 1 1 1 26
11 11 1 1 1 26
12 12 1 1 1 26
13 13 1 1 1 26
14 14 1 1 1 26
15 15 1 1 1 26
16 16 1 1 1 26
17 17 1 1 1 26
18 18 1 1 1 26
19 19 1 1 1 26
20 20 1 1 1 26
21 21 1 1 1 26
22 22 1 1 1 26
23 23 1 1 1 26
24  1 2 . .  .
25  2 2 1 2  3
26  1 3 . .  .
27  2 3 1 1 21
28  3 3 1 1 21
29  4 3 1 1 21
30  5 3 1 1 21
31  6 3 1 1 21
32  7 3 1 1 21
33  8 3 1 1 21
34  9 3 1 1 21
35 10 3 1 1 21
36 11 3 1 1 21
37 12 3 1 1 21
38 13 3 1 1 21
39 14 3 1 1 21
40 15 3 1 1 21
41  1 4 . .  .
42  2 4 0 2 22
43  3 4 0 2 22
44  4 4 0 2 22
45  5 4 0 2 22
46  6 4 0 2 22
47  7 4 0 2 22
48  8 4 0 2 22
49  1 5 . .  .
end
label values first_op type_of_op
label def type_of_op 1 "IRA", modify
label def type_of_op 2 "RPC", modify

------------------ copy up to and including the previous line ------------------

[Nick Cox]

Fortuitously, or otherwise, this question is being asked about once a day at the moment, so the advice please to read Statalist as well as write to it is pertinent. Otherwise https://www.stata.com/support/faqs/d...issing-values/ is one general discussion.

[Roshani Patel]

Ok thank you.... I managed to find it from a previous code you had provided
bysort id : replace gender = gender[_n+1] if missing(gender)

Thanks again for your swift response