1. The problem statement, all variables and given/known data Hi If I am dealing with an overdetermined system Ax=b, then I can (assuming A has full rank) find the unique approximative solution by least squares. Now, in my book it says that: "For a full column rank matrix, it is frequently the case that no solution x satisfies Ax=b exactly". I assume the book is saying that A having full rank is equivalent to it being overdetermined. Is that always the case? Best, Niles.