[R] SSbiexp

[Rui Barradas]

Hello,

Try the following.

fit <- nls(log(proc) ~ k + a*exp(b*cls), data = bline, start = c(k = 1, 
a = 1, b = 1))
summary(fit)
plot(fitted(fit), log(bline$proc))   # Seems nice


Hope this helps,

Rui Barradas

Em 23-04-2013 13:05, catalin roibu escreveu:
> Hello all!
> I have a problem to use a biexponential regression model:
> I use this code:
> n3<-nls(proc~SSbiexp(cls,a,b,c,d),data=bline) and this is the error message:
>
> Error in nls(y ~ cbind(exp(-exp(lrc1) * x), exp(-exp(lrc2) * x)), data =
> xy,  :
>    singular gradient
>
> My data is like this:
>
> structure(list(proc = c(387.177462830167, 508.090511433077,
> 321.836817656365,
> 151.226805860727, 150.885315536942, 86.2895998400175, 56.3215592398958,
> 39.5044440630984, 25.5703078997907, 7.33494872198676), cls = c(0.25,
> 0.5, 0.75, 1, 1.5, 2, 2.5, 3, 3.5, 4)), .Names = c("proc", "cls"
> ), row.names = c("0.25", "0.5", "0.75", "1", "1.5", "2", "2.5",
> "3", "3.5", "4"), class = "data.frame")
>
> Thank you!
>