Hello. In my example data, you'll see that most persons are unique to a different place.

However, there are a few instances where a person is in the data more than once, connected to different places. In these instances, I am hoping to drop the observation(s) where they have a lower -score- value (and keep all unique observations at the same time). I have been trying different bysort and collapse variations to no avail.

Thank you for your help!


Code:
* Example generated by -dataex-. To install: ssc install dataex
clear
input str5 place byte(person score)
"87343"  1 12
"87165"  2 14
"86987" 33 15
"86809" 16 11
"86631" 35 17
"86453"  6 27
"86275"  6 25
"83564" 18 24
"83386"  9 25
"83208" 10 26
"83030" 11 27
"82852" 11 15
"82674" 11 16
"82496" 14 17
"82318" 15 19
"82140" 15 13
"81962" 25  7
"81784" 24  8
"89000" 23 24
"88822" 29 22
end