Displaying 1 result from an estimated 1 matches for "initgamma".
2011 Oct 12
2
Nonlinear regression aborting due to error
...)
FIT <- nls(FORMULA, start=START, control=CONTROL, trace=T)
If GAMMA equals 10-80, nls converges successfully and the fit tracks the fit from a smoother (Supersmoother). However, if I attempt to estimate GAMMA using:
START <- list(EMAX=INITEMAX, EFFECT=INITEFFECT, C50=INITC50, GAMMA=INITGAMMA)
GAMMA increases rapidly to > 500 and nls terminates with:
Error in chol2inv(object$m$Rmat()) :
element (4, 4) is zero, so the inverse cannot be computed
In addition: Warning message:
In nls(FORMULA, start = START, control = CONTROL, trace = T) :
singular gradient
I also tried fixing...