Dear R-helpers,
I'm a new R-user and I was trying to gain some experience with the GNLS
function of the NLME package.
This is an extract from my dataset (it's a 432x6 data.frame) called "input",
in the first column I have the values that I need to fit, while the remaining
columns are input variables for the theoretical model, the function "mymodel"
(which returns a 432x1 vector of fitted values):
V0V1V2V3 V4 V5
0.56374863838.6875400.1041095890.0572495640
1.47392252639.125400.2767123290.0553929370
0.51704068441.0625450.1808219180.0532145910
1.38181318737.5625400.3534246580.0505213210
and this is my call of gnls:
fm1 <- gnls(V0 ~
mymodel(V1,V2,V3,V4,V5,par1,par2,par3,par4,par5,par6,par7,par8),
data=input,start=list(par1=0.2,par2=0.4,par3=0.8,par4=0.2,par5=10,par6=0.8,par7=0.9,par8=-0.5))
and I get the following error:
Error in `[.data.frame`(eval(model, data.frame(data, getParsGnls(plist, :
undefined columns selected
My guess is that R is not able to create that data.frame since "getParsGnls",
an internal function of gnls, does not provide its output (obviously the error
is mine, but I can't clearly identify it). Indeed, in the previous lines of the
gnls code, I don't have any problems when R creates a data.frame using the
dataframe "data" and "pars", as follows:
res <- eval(model, data.frame(data, pars))
I would be grateful to receive any help or hint.
Best regards,
Louis Salomon
[[alternative HTML version deleted]]
______________________________________________
[email protected] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
