[R] Problem with the port algorithm in nls function
roberto marrone
robertomarrone at hotmail.it
Mon Nov 16 19:30:08 CET 2015
Dear all,
using the following code I have find a problem with the port algorithm. If I use the nls function without lower bound for my parameter it compute the parameter's value but I want the parameter positive. Running the following code I had this error. Thanks at all.
optim <- nls(Prezzo ~ S*pnorm((log(15/14)+(0.015+theta^2/2)*0.17)/(theta*sqrt(0.17))) - 14*exp(-0.015*0-17)*pnorm((log(15/14)+(0.015+theta^2/2)*0.17)/(theta*sqrt(0.17)) - theta * sqrt(0.17)),
+ start=c(theta=0.1),
+ data=data,
+ algorithm="port",
+ lower=0.01,
+ trace=TRUE,
+ control= nls.control(maxiter = 100, tol = 1e-05, minFactor = 1/1024, printEval = FALSE, warnOnly = FALSE))
Error in nls_port_fit(m, start, lower, upper, control, trace, give.v = TRUE) :
INTEGER() can only be applied to a 'integer', not a 'NULL'
[[alternative HTML version deleted]]
More information about the R-help
mailing list