GLD.quantreg {GLDreg}

Fit a GLD quantile regression parametrically or non parametrically

Description

The GLD quantile regression can be: 1) Fixed intercept, allowing all other coefficients to vary, 2) Only intercept is allowed to vary and 3) All coefficients can vary. Minimisation is achieved numerically through least squares between the proportion of estimated GLD error distribution below zero versus the specified quantile for parametric approach. For non parametric approach, minimisation is achieved using a least squares approach to find a q-th quantile GLD line such that the percentage of observations below the line corresponds to the q-th quantile.

Usage

GLD.quantreg(q, fit.obj, intercept = "", slope = "", emp=FALSE)

Arguments

q

Specify the quantile (range 0 to 1) line

fit.obj

An object from GLD.lm.full

intercept

Can either be "fixed" or left blank, blank indicates this parameter is allowed to vary in quantile line estimation

slope

Can either be "fixed" or left blank, blank indicates this parameter is allowed to vary in quantile line estimation

emp

Can either be TRUE (non parametric GLD quantile regression) or FALSE (parametric GLD quantile regression), defaults to FALSE

Details

This is a wrapper function for fun.gld.all.vary, fun.gld.slope.fixed.int.vary, fun.gld.slope.vary.int.fixed.

Value

A matrix showing the estimated coefficients for the specified quantile regression model, the objective function value and whether convergence is reached in the optimisation process. A value of 0 indicates convergence is reached. The convergence value is the same as the one from the optim function.

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.full,fun.plot.q, 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)

## Find median regression, use empirical method

med.fit<-GLD.quantreg(0.5,fit,slope="fixed",emp=TRUE)

## Not run: 

## Extract the Engel dataset 

library(quantreg)
data(engel)

## Fit GLD Regression along with simulations

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

## Fit parametric GLD quantile regression from 0.1 to 0.9, with equal spacings 
## between quantiles

result<-GLD.quantreg(seq(0.1,.9,length=9),engel.fit.all,intercept="fixed")

## Non parametric quantile regression

GLD.quantreg(seq(0.1,.9,length=9),engel.fit.all,intercept="fixed",emp=T)


## End(Not run)