I am figuring out how to disentangle two unknown values using regressions, and I want to ask whether it is possible.
Let's say I have a measured value (y-axis) that is a sum of two unknowns (A and B). The sum of two unknowns has an underlying relationship with X, another measured value.
However, we know that A is constant, so it does not change with X, but B does vary.
If a linear regression satisfies A + B ~ X, is there a way to estimate A and B?
It might not be possible at all, but if we know that B varies with X, there might be a way around interpreting the intercept and the gradient to separate A and B...
This is a graph from what I am looking at: Relationship between A+B ~ X
The intercept is-0.08 and the gradient is 0.02
It would be extremely helpful if there are any suggestions for additional information required to separate A and B.
Thank you