M*********t
发帖数: 250 | 1 I'm not sure if variance is the right thing to look at,
I have: A+B+C = D,
1) it's pure linear transformation
2) some cases only has A and C, others only have B and C, only about 1/3 of
people have A, B, and C
I want to compare A, B, and C in terms of how much variance or percentage
they count in D.
I can't do correlation, because the squared correlation of AD, BD, and CD do
not add up to 1, because of missing cases
Is there a way that I can compare the relative importance/contribution of A,
B, and C to D? |
|
