Some Problems in Orthogonal Distance and Non-Orthogonal Distance Regression

摘要: Abstract : Of interest here is the problem of fitting a curve or surface to given data by minimizing some norm distances from points surface. These may be measured orthogonally surface, giving orthogonal distance regression, and for this problem, least squares has attracted most attention. Here we will look at two other important criteria, iota(sub 1)i Chebyshev norm. The former value when contain wild points, latter in context accept/reject criteria. There are however circumstances it not appropriate force orthogonal, possibilities also considered. first arises aligned with certain fixed directions, second angular information available about points. For norm, consider algorithmic developments these problems.