Would appreciate some help on this problem please:
Using nls to solve for 5 parameters to a set of data, and for particular
mixes of the parameters the function can yield NaN and infinity returns.
Is is making nls fall over I have tried starting values, lower and upper
bounds to no avail. New to R and not sure how to put constraints into this
function, the parameters have constraining relations.
Error messages I am getting with my tinkering are;
(1) Error in nlsModel(formula, mf, start, wts, upper) :
singular gradient matrix at initial parameter estimates
(2) NaNs produced
Thanks, any help for a newbie would be great.
Glenn
[[alternative HTML version deleted]]