GLD.lm.full {GLDreg}

This function fits a GLD regression linear model and conducts simulations to display the statistical properties of estimated coefficients

Description

The function is an extension of GLD.lm and defaults to 1000 simulation runs, coefficients and statistical properties of coefficients can be plotted as part of the output.

Usage

GLD.lm.full(formula, data, param, maxit = 20000, fun, method = "Nelder-Mead", 
range = c(0.01, 0.99), n.simu = 1000, summary.plot = TRUE, init = NULL)

Arguments

formula

A symbolic expression of the model to be fitted, similar to the formula argument in lm, see formula for more information

data

Dataset containing variables of the model

param

Can be "rs", "fmkl" or "fkml"

maxit

Maximum number of iterations for numerical optimisation

fun

If param="fmkl" or "fkml", this can be one of fun.RMFMKL.ml.m, fun.RMFMKL.ml, for maximum likelihood estimation (*.ml.m is a faster implementation of *.ml) and fun.RMFMKL.lm for L moment matching.

If param="rs", this can be one of fun.RPRS.ml.m, fun.RPRS.ml, for maximum likelihood estimation (*.ml.m is a faster implementation of *.ml) and fun.RPRS.lm for L moment matching.

method

Defaults to "Nelder-Mead" algorithm, can also be "SANN" but this is a lot slower and may not as good

range

The is the quantile range to plot the QQ plot, defaults to 0.01 and 0.99 to avoid potential problems with extreme values of GLD which might be -Inf or Inf.

n.simu

Number of times to repeat the simulation runs, defaults to 1000.

summary.plot

Whether to plot the coefficients graphically, defaults to TRUE.

init

Choose a different set of initial values to start the optimisation process. This can either be full set of parameters including GLD parameter estimates, or it can just be the coefficient estimates of the regression model.

Details

This function usually takes some time to run, as it involves refitting the GLD regression model many times, the progress of the simulation is outputted to the R screen, so users can guage the progress of the computation.

Value

[[1]]

Output of GLD.lm

[[2]]

A matrix showing the bias adjustment, coefficents of the model, parameters of GLD and whether the result converged at each run

[[3]]

Adjusted simulation result so that the empirical mean of coefficients is the same as the estimated parameters obtained in GLD.lm

Author(s)

Steve Su

References

Su (2015) "Flexible Parametric Quantile Regression Model" Statistics & Computing May 2015, Volume 25, Issue 3, pp 635-650

See Also

GLD.lm, GLD.quantreg, summaryGraphics.gld.lm

Examples

## Dummy example

## Create dataset

set.seed(10)

x<-rnorm(200,3,2)
y<-3*x+rnorm(200)

dat<-data.frame(y,x)

## Fit FKML GLD regression with 3 simulations

fit<-GLD.lm.full(y~x,data=dat,fun=fun.RMFMKL.ml.m,param="fkml",n.simu=3)

## Not run: 
## Extract the Engel dataset 

library(quantreg)
data(engel)

## Fit a full GLD regression

engel.fit.full<-GLD.lm.full(foodexp~income,data=engel,param="fmkl",
fun=fun.RMFMKL.ml.m)

## Extract the mammals dataset 
library(MASS)

## Fit a full GLD regression

mammals.fit.full<-GLD.lm.full(log(brain)~log(body),data=mammals,param="fmkl",
fun=fun.RMFMKL.ml.m)

## Using quantile regression coefficients as starting values
library(quantreg)

mammals.fit1.full<-GLD.lm.full(log(brain)~log(body),data=mammals,param="fmkl",
fun=fun.RMFMKL.ml.m, init=rq(log(brain)~log(body),data=mammals)$coeff)

## Using the result of mammals.fit.full as initial values

mammals.fit2.full<-GLD.lm.full(log(brain)~log(body),data=mammals,param="fmkl",
fun=fun.RMFMKL.ml.m, init=mammals.fit1.full[[1]][[3]])

## End(Not run)