I would like to do a distribution analysis with below data.
Sale Record:
Name, Month, SaleAmt
sale A, 201501, 200
sale A, 201501, 100
sale A, 201502, 500
sale A, 201503, 400
sale B, 201502, 300
sale B, 201502, 50
sale B, 201503, 400
sale C, 201503, 300
Monthly Target (per person):
201501, 200
201502, 300
201503, 350
I would like to do a distribution analysis compare the total actual sales per person versus the monthly target (in terms of percentage).
With aggr function, I think I can build a virtual table like
Name, Month, SaleAmt
sale A, 201501, 300
sale A, 201502, 500
sale A, 201503, 400
sale B, 201502, 350
sale B, 201503, 400
sale C, 201503, 300
To further analysis, since the start month of these three sales are in different months, so the figures should be sth like:
For Sale A:
(300 + 500 + 400) / (200 + 300 + 350)
For Sale B:
(350+ 400) / (300 + 350)
For Sale C:
(300) / (350)
And based on these figures, I should then able to create a Grid chart to analysis the distribution (e.g. count of sale by variance in sale target).
However, I have no concrete ideas to do the aforesaid analysis except the function aggr / class should be used. Any idea are welcomed.
