Hi there are some predefined self-start functions, like SSmicmen, SSbiexp, SSasymp, SSasympOff, SSasympOrig, SSgompertz, SSflp, SSlogis, SSweibull, Quadratic, Qubic, SSexp (nlrwr) Btw, do you know graphic examples for this functions? The SSexpDecay (exponential decay) for y = (y0 - plateau)*exp(-k*x) + plateau from http://stats.stackexchange.com/questions/17126/how-can-i-make-my-r-nls-model-more-stable works fine, too. But now I search a self-starter for y = a + b * c^x Do you know a package that implement such self-starter or is there a easy way to do it myself? thx Christof
Christof Klu? <ckluss <at> email.uni-kiel.de> writes:> But now I search a self-starter for y = a + b * c^x > > Do you know a package that implement such self-starter or is there a > easy way to do it myself?You might be able to adapt the SS.calib function from the psyphy package for this. It was designed to fit gamma functions to the luminance vs frame buffer values measured on CRT screens. But the functional form is similar.> thx > Christof > >best, Ken -- Kenneth Knoblauch Inserm U846 Stem-cell and Brain Research Institute Department of Integrative Neurosciences 18 avenue du Doyen L?pine 69500 Bron France tel: +33 (0)4 72 91 34 77 fax: +33 (0)4 72 91 34 61 portable: +33 (0)6 84 10 64 10 http://www.sbri.fr/members/kenneth-knoblauch.html
Hi Ken Am 14-08-2012 11:50, schrieb ken knoblauch:> Christof Klu? <ckluss <at> email.uni-kiel.de> writes: > >> But now I search a self-starter for y = a + b * c^x > > You might be able to adapt the SS.calib function > from the psyphy package for this. > It was designed to fit gamma functions to the > luminance vs frame buffer values measured on CRT > screens. But the functional form is similar.thank you for the hint. I will test that function too. But it seems to be something like y = a + b * x^c and not y = a + b * c^x Greetings Christof
On 2012-08-14 00:09, Christof Klu? wrote:> Hi > > there are some predefined self-start functions, like SSmicmen, SSbiexp, > SSasymp, SSasympOff, SSasympOrig, SSgompertz, SSflp, SSlogis, SSweibull, > Quadratic, Qubic, SSexp (nlrwr) > > Btw, do you know graphic examples for this functions? > > The SSexpDecay (exponential decay) for y = (y0 - plateau)*exp(-k*x) + > plateau from > http://stats.stackexchange.com/questions/17126/how-can-i-make-my-r-nls-model-more-stable > works fine, too. > > But now I search a self-starter for y = a + b * c^x > > Do you know a package that implement such self-starter or is there a > easy way to do it myself?Can't you just reparameterize with c^x = exp(x * log(c)) ? Peter Ehlers> > thx > Christof