Hi guys,

I'm trying to collapse a large dataset by year, gender, and economic sector but instead of the usual percentile groups I'm trying to use quintiles, or deciles.

I know that xtile is a popular command but it doesn't work with by or with bysort , and I'm splitting things by three dimensions. I outline an example using the unions dataset below.

Code:
webuse union , clear tab1 year union black age

I know how to make this work for a given percentile, so for the tenth percentile I could write this

Code:
collapse /// (p10) age (count) N=age, /// by(year union black)
The only problem is I want to scale up from percentiles to deciles or quintiles. Something which would look kind of like this.

Code:
collapse /// (p1-p10) age (count) N=age, /// by(year union black)
Does anyone have any ideas? Any help would be appreciated.