How to compute the sum of the first occurence within Group(s) in Stata?

[smith Jason]

I have a small long format dataset below.
* Example generated by -dataex-. For more info, type help dataex
clear
input byte (gender period pr)
1 1 11
1 1 11
1 1 11
1 2 15
1 2 15
1 2 15
2 1 11
2 1 11
2 1 11
2 2 12
2 2 12
2 2 12
end

I want to create a new variable sum_pr,
which equals to the sum of the first value within gender and period like below:
* Example generated by -dataex-. For more info, type help dataex
clear
input byte (gender period pr sum_pr)
1 1 11 26
1 1 11 26
1 1 11 26
1 2 15 26
1 2 15 26
1 2 15 26
2 1 11 23
2 1 11 23
2 1 11 23
2 2 12 23
2 2 12 23
2 2 12 23
end

How can I use Stata code to achieve it?
Thank you!

[William Lisowski]

 generate seq = _n

. bysort gender (period seq): egen sum_pr = total(cond(period!=period[_n-1],pr,0))

. drop seq

. list, sepby(gender)

     +-------------------------------+
     | gender   period   pr   sum_pr |
     |-------------------------------|
  1. |      1        1   11       26 |
  2. |      1        1   11       26 |
  3. |      1        1   11       26 |
  4. |      1        2   15       26 |
  5. |      1        2   15       26 |
  6. |      1        2   15       26 |
     |-------------------------------|
  7. |      2        1   11       23 |
  8. |      2        1   11       23 |
  9. |      2        1   11       23 |
 10. |      2        2   12       23 |
 11. |      2        2   12       23 |
 12. |      2        2   12       23 |
     +-------------------------------+

[smith Jason]

Thank you!

[smith Jason]

I have to bother your guys again. Sorry, I think I didn't express my thoughts very well.
What I really want is listed as follows, and it should be running sum instead of sum.
* Example generated by -dataex-. For more info, type help dataex
clear
input byte (gender period pr sum_pr)
1 1 11 11
1 1 11 11
1 1 11 11
1 2 15 26
1 2 15 26
1 2 15 26
2 1 11 11
2 1 11 11
2 1 11 11
2 2 12 23
2 2 12 23
2 2 12 23
end

Can someone keep helping me in Stata?
Thank you!

[Leonardo Guizzetti]

Sorry, no, not the way you have posted this data. Please take a moment to read the output of -help dataex- and post code and output within the code delimiters. This will make it easy for us to read and, if necessary, copy data into Stata.

[smith Jason]

If I did something wrong, please forgive me and I have to open a new post here to ask for help.

[Leonardo Guizzetti]

Thank you for editing your post in #4 to make the dataset usable after dataex. It can be a bit tricky to ask questions here and get the formatting right, etc. No need to apologize, it's a learning curve.

This will get you a running sum by gender and period.

clear
input byte (gender period pr sum_pr)
1 1 11 11
1 1 11 11
1 1 11 11
1 2 15 26
1 2 15 26
1 2 15 26
2 1 11 11
2 1 11 11
2 1 11 11
2 2 12 23
2 2 12 23
2 2 12 23
end

egen first = tag(gender period)
by gender (period), sort: gen want = sum(pr) if first
by gender period, sort: replace want = want[1] if mi(want)
list, sepby(gender period)

Result

. list, sepby(gender period)

     +----------------------------------------------+
     | gender   period   pr   sum_pr   first   want |
     |----------------------------------------------|
  1. |      1        1   11       11       1     11 |
  2. |      1        1   11       11       0     11 |
  3. |      1        1   11       11       0     11 |
     |----------------------------------------------|
  4. |      1        2   15       26       1     26 |
  5. |      1        2   15       26       0     26 |
  6. |      1        2   15       26       0     26 |
     |----------------------------------------------|
  7. |      2        1   11       11       1     11 |
  8. |      2        1   11       11       0     11 |
  9. |      2        1   11       11       0     11 |
     |----------------------------------------------|
 10. |      2        2   12       23       1     23 |
 11. |      2        2   12       23       0     23 |
 12. |      2        2   12       23       0     23 |
     +----------------------------------------------+

[smith Jason]

Thank you very much!

[Leonardo Guizzetti]

You're welcome.