wilnl: Computes the Rank-Based Fit of a Nonlinear Model

Description Usage Arguments Author(s) References

Description

Computes the rank-based fit of a nonlinear model, either the Wilcoxon (“WIL”) or HBR fit. The Wilcoxon fit is discussed in Chapter 3 of Hettmansperger and McKean (2011) and the HBR fit is developed in Abebe and McKean (2014). See Section 7.7 of Kloke and McKean (2014) for a discussion of the Rfit version.

Usage

1
wilnl(x, y, theta0, fmodel, jmodel, numstp = 50, eps = 0.001, wts.type = "WIL", intest = "HL", intercept = FALSE)

Arguments

x

matrix of predictors

y

response vector

theta0

initial estimate of nonlinear parameters

fmodel

R function for the model

jmodel

R function for the Jacobian

numstp

maximum number of iterative steps (default is 50)

eps

precision tolerance (default is 0.001)

wts.type

either "WIL" (default) for the Wilcoxon fit or "HBR" for the HBR fit

intest

either "HL" (default) for Hodges-Lehmann estimator of the intercept or "MED" for the median estimator

intercept

TRUE if an intercept is in the model else FALSE

Author(s)

Joe McKean mckean@wmich.edu and John Kloke kloke@biostat.wisc.edu

References

Abebe, A. and McKean, J.W. (2014), Weighted Wilcoxon estimators in nonlinear regression, Australian and New Zealand Journal of Statistics, 55, 401-420.

Hettmansperger, T.P. and McKean J.W. (2011), Robust Nonparametric Statistical Methods, 2nd ed., New York: Chapman-Hall.

Kloke, J. and McKean, J.W. (2014), Nonparametric statistical methods using R, Boca Raton, FL: Chapman-Hall.


kloke/npsmReg2 documentation built on May 20, 2019, 12:34 p.m.