The nls.lm function provides an R interface to lmder and lmdif from the MINPACK library, for solving nonlinear least-squares problems by a modification of the Levenberg-Marquardt algorithm, with support for lower and upper parameter bounds. The implementation can be used via nls-like calls using the nlsLM function.
Version 1.2-1 o In some cases 'profile.nls' would not work on models created with 'nlsLM', due to a wrong class assignment in the output ("call" instead of "numeric"). This bug was kindly made aware by Patrice Kiener and removed. o License is now only GPL-3; per clause 7 we include the University of Chicago license terms as a supplement.
Version 1.2-0 o Removed a small bug in 'nlsLM' that gave an error if an external function was supplied within the formula syntax, e.g. y ~ fitModel(x, a). This was made aware by an email from C. Softley, pointing to a discussion and solution in http://stackoverflow.com/questions/30089698/error-in-evalexpr-envir-enclos-could-not-find-function-nested-functions.
o Modified NAMESPACE and added Suggests: MASS to DESCRIPTION file.
o The new function wfct for easy specification of weights in nls and
nlsLM is now included.
Version 1.1-7 o changed the License to GPL-3 + the terms specified in the file LICENSE, which give the original copyright distributed with
the minpack Fortran source. This is possible given Clause 7 in GPL-3.
Version 1.1-6 o Support for lower and upper bounds added o Added 'nlsLM', a modified version of 'nls' that uses ‘nls.lm’ for fitting.