Usage
nnr.control(job = -1, tol = 0, max.iter = 50, init = 0, limnla = c(-10,
0), varht = NULL, theta = NULL, prec = 1e-06, maxit = 30,
method = "NR", increment = 1e-04, backfit = 5, converg = "coef",
toler = 0.001)Arguments
job
an integer representing the optimization method used to find the smoothing parameter.
The options are job=-1: golden-section search on (limnla(1), limnla(2));
job=0: golden-section search with interval specified automatically;
job >0: regular grid sear
tol
tolerance for truncation used in `dsidr'. Default is 0.0, which sets to square of machine precision.
max.iter
maximum number of iterations allowed for the Gauss-Newton/Newton-Raphson iteration.
init
an integer of 0 or 1 indicating if initial values are provided for theta. If init=1, initial values are provided using theta. Default is 0.
limnla
a vector of length 2, specifying a search range for the n times smoothing parameter on log10 scale. Default is (-10, 0).
varht
needed only when vmu="u", which gives the fixed variance in calculation of the UBR function. Default is NULL.
theta
If `init=1', theta includes intial values for smoothing parameters. Default is NULL.
prec
precision requested for the minimum score value, where precision is the weaker of the absolute and relative precisions. Default is 1e-06.
maxit
maximum number of iterations allowed. Default is 30.
method
a character string specifying a method for iterations, "GN" for Gauss-Newton and "NR" for Newton-Raphson. Default is "GN".
backfit
an integer representing the number of backfitting iterations for multiple functions. Default is 5.
converg
an optional character, with possible values "coef" and "ortho", specifying the convergence
criterion to be used. "coef" uses the change of estimate of parameters and functions to
assess convergence, and "ortho" uses a criterion similar to the relative
toler
tolerance for convergence of the algorithm. Default is 0.001.