On Tuesday, August 12, 2014 10:10:12 PM UTC+12, Aino wrote:
> The bigger picture (for those who are interested): I have two sets of data and two regression lines, and I have to decide to which data set the sample belongs to. Linear discriminant analysis is not an option here
It sounds like you start with 2 sets of data both consisting of (X,Y) pairs. Then you get a new (X,Y) pair and want to know whether it is more likely to come from set 1 or set 2. Linear discriminant analysis seems perfect for this problem, at least if the assumptions for linear regression are met. So, I'm curious why you said that discriminant analysis is not an option?