removing duplicates based on a criteria

[Khushbakht Jamal]

Hello, I have a dataset with village names and their respective populations. Villages belong to different districts. The data structure looks like below:

Village_name	District	pop
shahpur	A	10
ali pur	B	20
chak 15	B	30
chak 15	B	40
chak 20	C	50

I want to get a unique dataset by dropping the duplicate villages. But while dropping, I want to keep the village name with higher population, in the example above I want to keep chak 15 with 40 people and not 30.

How can I do so? Please let me know.

Thanks.

[Hemanshu Kumar]

Welcome to Statalist, Khushbakht!

Something like this should do it:

bysort District Village_name (pop): keep if _n == _N

warning: this is untested code, since a data example was not properly posted. Please go through Statalist FAQ, especially section 12, to see how data examples should be posted (using the dataex command in Stata).