Hi, I am looking for an more automated way to fit functions of the form y~a*x^b+c to given data, for example x<-c(0.36,0.63,0.90,1.166,1.43, 1.70, 1.96, 2.23) y<-c(9.0,9.62,10.11,10.53,10.9, 11.25, 11.56, 11.86) I tried for example with nls, which did only work with really good initial guessed values. The idea is to use a function, which would not require to guess c in order to linearize the equation. Thanks a lot Thomas --------------------------------- Thomas Bschorr Department of Physics Institute of Astronomy ETH Zurich Wolfgang-Pauli-Strasse 27, HIT J 11.3 CH-8093 Zürich Switzerland E-mail: bschorr@phys.ethz.ch Phone: +41 (0)44 633-6684 Fax: +41 (0)44 633-1238 Web: http://www.exp-astro.phys.ethz.ch <http://www.exp-astro.phys.ethz.ch/> [[alternative HTML version deleted]]