dfr.corrts: Derivative free Two Stage estimate

In nlr: Nonlinear Regression Modelling using Robust Methods

Description

Derivative free two stage estimate for nonlinear regreession model with autocorrelated error.

Usage

dfr.corrts(formula, data, start = getInitial(formula, data), 
control = nlr.control(tolerance = 0.001, minlanda = 1/2^10, 
maxiter = 25 * length(start)), correlation = 1, ...)

Arguments

formula	nl.form object of the nonlinear function model. See nl.form object.
data	list of data with the response and predictor as name of variable.
start	list of starting value parameter, name of parameters must be represented as names of variable in the list.
control	nlr.control object, include tolerance, maxiter,... see nlr.control.
correlation	correlation structure, at the moment parameter of AR(p) process.
...	any argument pass to formula

Details

In first stage nonlinear regression parameter estimate and in second stage autocorrelation structure estimate and finally the generalized least square estimates the function model parameters.

In this function all stages compute by derivative free methods, which minimization methods uses Nelder-Mead method.

Value

fited	nl.fitt.gn object generated by nlsnm function.
tm	fitted time series model for residuals.

Note

This function currently run with AR process. The robust estimate is don by nl.robcorrts function. This function will be called from nlr by providing correlation as correlation structure and derivfree. It is under developement and for internal use, user might not call it directly, it is more efficient to call from nlr function with mentioned arguments.

Author(s)

Hossein Riazoshams, Jul 2009. Email: [e-mail] URL http://www.riazoshams.com/nlr/

References

Riazoshams, H., Midi, H., Sharipov, O. S.H, (2010). The Performance of Robust Two Stage Estimator in Nonlinear Regression with autocorrelated Error, Communications in Statistics - Simulation and Computation, 39: 1251-1268.

See Also

nl.robcorrts, nlsqr.gn, nl.fitt.gn, nlr.control, nlsnm

Examples

# The direct call of nlr call dfr.corrts.
p1<- 8.06e+10
p2<- 1e11
p3<-1970
p4=6
chstart2 <- list(p1=p1,p2=p2,p3=p3,p4=p4)
irandt<-nlr::trade.ir
dfrir<- dfr.corrts (nlrobj5[[4]],data=list(xr=irandt[,1],yr=irandt[,2]),start=chstart2,
control=nlr.control(trace=TRUE),correlation = 2)
dfrir$fited$parameters

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