nnr.control: Set Control Parameters for nnr

In catherinewang1/assist: A Suite of R Functions Implementing Spline Smoothing Techniques

Description

Control parameters supplied in the function call replace the defaults to be used in calling nnr.

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 search on [limnla(1), limnla(2)] with \#(grids) = job + 1. Default is -1.
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".
increment	specifies a small value as increment to calcuate derivatives. Default is 1e-04.
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 offset used in nls. Default is "coef".
toler	tolerance for convergence of the algorithm. Default is 0.001.

Value

returned is a list includes all re-seted control parameters.

Author(s)

Chunlei Ke chunlei\[e-mail] and Yuedong Wang [e-mail]

See Also

nnr, dsidr,dmudr

Examples

## Not run: 
## use Newton-Raphson 
nnr.control(method="NR")

## End(Not run)

catherinewang1/assist documentation built on June 16, 2019, 1:36 p.m.