Dear Forum,
I have the following very basic problem. Unfortunately, I couldn't find an appropriate solution in the forum. so please correct me in case there`s a thread out there already.

My dataset is structured as follows:
ID year industry state X Y
2 2010 0K70 1 1.9e+07 1.4e+09
2 2011 0K70 1 3.7e+07 1.7e+09
6 2010 0K70 1 2.5e+06 3.5e+07
6 2011 0K70 1 2.5e+06 4.3e+07
8 2010 0F51 0 270000
8 2011 0F51 0 624000
9 2010 0S90 0 4.8e+06 3.2e+07
9 2011 0S90 0 5.1e+06 2.1e+07

I want to calculate the variable Z constructs a pay gap benchmark using "non-state" observations. which Specifically shows the average pay gap of non-state in each industry-year, and then compares the "state" pay gap in the same industry year with the non-state industry-year average.
( Z ) is calculated using the following formula :
Z = average_ Non_state_paygap j,t - state paygap i,j,t

where: Average Pay Gap_Non_SOEj,t is the average pay gap calculated for industry j in year t using Non-state. PayGap is the ratio of the average X to the average Y, and it is calculated for state and non-state firm-year observations, respectively.
Any help would be highly appreciated, thank you in advance.

omer