Estimating Missing Data
Not sure if this is the right forum, or site for that matter, but hopefully someone can help me out.
I have three sets of data which includes three subsets. Lets say x, y and z. z is made up entirely of data from both x and y, however I do not know how much of z is made up of x and how much is made up of y. Sets x, y and z, contain a subset each. I'll call them x1, y1, z1. I know the percentage of x that is x1, y that is y1 and z that is z1 and z1 is also made up entirely of x1 and y1.
So to break it down.
x = 1000 units
y = 500 units
z = 40 units
x1 = 20 (2% of x)
y1 = 75 (15% of y)
z1 = 4 (10% of z)
So my question is, is it possible to estimate what value of z is made up of x and what value of z is made up of y knowing the percentages of x1, y1 and z1.
I think it can be done by working out what amount x1 is of z1 and y1 is of z1 then applying that to x and y.