Module DODRC in CMLIBComputes the weighted orthogonal distance regression or ordinary linear or nonlinear least squares solution. Derivatives are either supplied by the user or numerically approximated. Control values are supplied by the user, and a three-part report of the results is optionally generated.
CALL DODRC(FUN, JAC, N, M, NP, X, LDX, IFIXX, LDIFX, SCLD, LDSCLD, Y, BETA, IFIXB, SCLB, RHO, LDRHO, W, JOB, TAUFAC, SSTOL, PARTOL, MAXIT, IPRINT, LUNERR, LUNRPT, WORK, LWORK, IWORK, LIWORK, INFO)| K1a1a3 | Unconstrained linear least squares approximation on univariate data (curve fitting) |
| K1b1a1 | Unconstrained nonlinear least squares approximation by smooth functions, user provides no derivatives |
| K1b1a2 | Unconstrained nonlinear least squares approximation by smooth functions, user provides first derivatives |
| L8a4 | Simple linear errors in variables regression |
| L8c5 | Measurement error models |
| L8e5 | Nonlinear measurement error models |