I would like to know the link between general position and multicollinearity. In statistics, a multivariate dataset of dimension p is said to be in general position if at most p observations lies in a (p-1) dimensional hyperplane. That is, for a 3 dimensional data-set, there must be at most 3 observations on a plane.
That is, for a 3x3 square matrix, there might be at most 3 points in the same plane. Is that similar to saying that this 3x3 matrix has full-rank ? If the multivariate datasets is not in general position, does it imply multicollinearity ?