How to generate new variables based on previous year observations

[Ke Gong]

Dear experts,

I have a data example shown below. This dataset consists of multiple year firm revenue observations. What I need is to generate two new variables "pre2012" and "post2012". "pre2012" will be the average revenue of year 2010-2012 which is 60, while "post2012" will be the average revenue of year 2013-2015 which is 85. Could you please show me how to create these two variables? Than you so much!

* Example    generated by -dataex-.    To install: ssc install    dataex
clear
year    firm    revenue      pre2012     post2012
2010    1          50           60         85
2011    1          60           60         85
2012    1          70           60         85
2013    1          80           60         85
2014    1          85           60         85
2015    1          90           60         85
end

[Clyde Schechter]

* Example generated by -dataex-. To install: ssc install dataex
clear
input float(year firm revenue)
2010 1 50
2011 1 60
2012 1 70
2013 1 80
2014 1 85
2015 1 90
end

by firm, sort: egen pre2012 = mean(cond(inrange(year, 2010, 2012), revenue, .))
by firm: egen post2012 = mean(cond(inrange(year, 2013, 2015), revenue, .))

list, noobs clean

[Ke Gong]

Clyde Schechter Thank you so much! It works pretty well.