nl.lmsGA: Fitt a nonlinear regression model by least median of squares....

In nlr: Nonlinear Regression Modelling using Robust Methods

Description

Least Median of square estimate is robust fitt by minimizing the median of squared residuals. This function use the "ga" function,from "GA" package, which minimize using genetic algorithm method.

Usage

nl.lmsGA(formula, data, start,min=NULL,max=NULL,type="real-valued")

Arguments

formula	nl.form object of the nonlinear function model.
data	list of data include responce and predictor.
start	list of parameter values of nonlinear model function (θ. in f(x,θ)).
min	vector of minimum values of parameters, which passes to "ga" function.
max	vector of maximum values of parameters, which passes to "ga" function.
type	the type of genetic algorithm to be run, which passes to "ga" function.

Details

Note that due to using genetic algorithm method, this function is mor efficient than nl.lmsNM, but estimator due to non uniqueness of minimum of objective function is not efficient, but is high breakdown estimate.

Value

list of parameter estimates.

Note

When that function is not working properly, nl.lmsNM function can be used, but requires more precise initial values. These function uses by nlr for initial purposes. User can request least median square as initial value from nlr.control argument by initials="lms" method, nlr use nl.lmsGA not nl.lmsNM.

Author(s)

Hossein Riazoshams, May 2014. Email: [e-mail] URL http://www.riazoshams.com/nlr/

References

Riazoshams H, Midi H, and Ghilagaber G, 2018,. Robust Nonlinear Regression, with Application using R, Joh Wiley and Sons.

See Also

nl.lmsNM, nlr.control, nlr

Examples

  # chicken data fitt example
  data=list(xr=Weights$Date,yr=Weights$Weight)
  fit<- nl.lmsGA(nlrobj1[[14]],data=data,start=list(p1=100,p2=42,p3=.11))

nlr documentation built on July 31, 2019, 5:09 p.m.