Center for Computing and Applied Mathematics

摘要: ODRPACK is a software package for weighted orthogonal distance regression, ie, for nding the parameters that minimize the sum of the squared weighted orthogonal distances from a set of observations to the curve or surface determined by the parameters. It can also be used to solve the nonlinear ordinary least squares problem. The procedure has application to curve and surface tting, and to measurement error models in statistics. ODRPACK can handle both explicit and implicit models, and will easily accommodate complex and other types of multiresponse data. The algorithm implemented is an e cient and stable trust region Levenberg-Marquardt procedure that exploits the structure of the problem so that the computational cost per iteration is equal to that for the same type of algorithm applied to the nonlinear ordinary least squares problem. The package allows a general weighting scheme, provides for nite di …