You may call the code of the function "nls" and save it in a personal function, say "fu.nls". Then look at the code; at the end, there is an instruction : if(z$control[5]) stop(switch(as.integer(z$co
I have a factorial design to analyse (5 doses, 2 times, 3 batches, and replications of value). The design is as follows: Dose<o:p></o:p> Dose<o:p></o:p> Dose<o:p></o:p> Dose<o:p></o:p> Dose<o:p></o:
I was recently driven to use a SAS program and an (apparently) similar S+ program on the same data, and I got rather different results. As I trust both SAS and S+, the only solution is a problem in
Dear S+ users, There is something I don't understand about S+ function "nls" and any advice is welcome. Here is the problem : when I run "nls" in a program I made to fit some data with a non-linear c
Dear S+ users, I am struggling with how contrasts can be calculated in S+. For the sake of simplicity, I will take a short example. Suppose a dataset X containing factor variables "A", "B", "C" and a
Dear all specialist users of S+, I had to use the following code : data = W) to analyse the so-called "Value", with "TimeNumber" refering to a repeated measurement design (2 different days of evaluat
I have an array, say "W", arranged like that : X1 X2 X3 Y1 Y2 Y3 1 41893 12803 12803 12202 121847 96500 2 133033 302643 302643 40409 149891 62261 3 (...) I was wondering how to apply a wilcoxon test
I am wondering whether it is possible to easily test the goodness-of-fit of a distribution curve against a normal distribution curve. For example, I have a set of data, say X, and using qqnorm(X) an
Thanks to many people for giving me an answer to the following question I posted last Wednesday : "I am wondering whether it is possible to easily test the goodness-of-fit of a distribution curve aga
I have some "philosophical" problems using the S+ function "lme". Basically, the data I want to analyze are presented like this : Spot Gel Donor Type Value 1 81 O2 1 Cy3 4.626139 244 81 04 1 Cy3 3.9
Sorry, I sent a message with a mistake within : the "%*%" expression doesn't exist... I wanted to say "%in%"... It works, but with a "non convergence" warning. Nevertheless, the question remains. Tha
To whom can help : I made a program in which I am using the S+ "nls" function to estimate some parameters of a given formula. Globally : estimation <- nls( ... parameters passed to the function...) e