Dear chenu, I am not going to see your code with attention (also I do not understand the `*' symbol you used), however it looks a changepoint-type problem.
The package segmented (on CRAN) uses a piecewise linear parameterization to fit regression models with breakpoints. Hope this helps, best, vito ----- Original Message ----- From: chenu <[EMAIL PROTECTED]> To: <[EMAIL PROTECTED]> Sent: Monday, December 15, 2003 11:01 AM Subject: [R] nls arguments > Hi all, > > I've got a problem with the nls function. > I have an adjustment which works when I fix one of the argument of my > function (Xo=150) : > > *Xo*=150 > f<- function (tt*,Xo*,a,b) ifelse(tt<*Xo*,a*exp(-b**Xo*),a*exp(-b*tt)) > ajust<-nls(RER~f(tt,*Xo*,a,b),data=data.frame(tt=Ph2[,2*k],RER=Ph2[,2*k+1]), start=list(a=0.5,b=0.014)) > > But, when I use it as a "normal" parameter (and even if I do not I use > it in the equation of my "f" function), it does not work : > > f<- function (tt,*Xo*,a,b) ifelse(tt<150,a*exp(-b*tt),a*exp(-b*tt)) > or : f<- function (tt,*Xo*,a,b) > ifelse(tt<*Xo*,a*exp(-b**Xo*),a*exp(-b*tt)) #it is this > equation I would like to resole > ajust<-nls(RER~f(tt,*Xo*,a,b),data=data.frame(tt=Ph2[,2*k],RER=Ph2[,2*k+1]), start=list(a=0.5,b=0.014,*Xo*=150)) > /Error in nlsModel(formula, mf, start) : singular gradient matrix at > initial parameter estimates/ > > > > Would you have any idea to help me. > > Thanks a lot > > Karine Chenu > > ______________________________________________ > [EMAIL PROTECTED] mailing list > https://www.stat.math.ethz.ch/mailman/listinfo/r-help ______________________________________________ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help