Generate a new variable and assigns 1 after a value increases in other variable.

[Ryan Kim]

Dear Statalist users,

I've been still struggling for a few hours after I tried several commands searched from the forums.
What I want to do is I want to create a new variable and assigns 1 if there is a value increase in old_var.
In other words, 1 needs to be continuously assigned after 3rd observation.

I also want to apply an another condition that enables me to keep assigning 1 until the last observation even if there is a decrease/increase in the old_var at 8th/10th observation.

To conclude, is there any way that I can keep assigning 1 only after the first value change (from 0 to 1) in old_var and still assigns 1 even if there is a value change later?

Below is the example data set and the expected output.
Thank you very much in advance.

* Example generated by -dataex-. To install: ssc install dataex
clear
input str1 id float(time old_var)
"a"  1 0
"a"  2 0
"a"  3 1
"a"  5 1
"a"  6 1
"a"  7 1
"a"  9 1
"a" 10 0
"a" 11 0
"a" 12 1
end

                    
     +-------------------------------+
     | id   time   old_var   new_var |
     |-------------------------------|
  1. |  a      1         0         0 |
  2. |  a      2         0         0 |
  3. |  a      3         1         1 |
  4. |  a      5         1         1 |
  5. |  a      6         1         1 |
     |-------------------------------|
  6. |  a      7         1         1 |
  7. |  a      9         1         1 |
  8. |  a     10         0         1 |
  9. |  a     11         0         1 |
 10. |  a     12         1         1 |
     +-------------------------------+

[Clyde Schechter]

You do not say as much, but I assume you want to do this starting over at 0 for each new id.

by id (time), sort: gen new_var = min(sum(old_var != old_var[_n-1]) - 1, 1)