lqm.fit.gs: Quantile Regression Fitting by Gradient Search
In lqmm: Linear Quantile Mixed Models

Description Usage Arguments Details Value Author(s) References See Also Examples

This function controls the arguments to be passed to routines written in C for LQM estimation. The optimization algorithm is based on the gradient of the Laplace log–likelihood (Bottai, Orsini and Geraci, 2013).

1	lqm.fit.gs(theta, x, y, weights, tau, control)

`theta`	starting values for the regression coefficients.
`x`	the model matrix.
`y`	the model response.
`weights`	the weights used in the fitting process.
`tau`	the quantile to be estimated.
`control`	list of control parameters used for optimization (see `lqmControl`).

See argument fit in lqm for generating a list of arguments to be called by this function.

An object of class list containing the following components:

`theta`	a vector of coefficients.
`scale`	the scale parameter.
`gradient`	the gradient.
`logLik`	the log–likelihood.
`opt`	number of iterations when the estimation algorithm stopped.

Marco Geraci

Bottai M, Orsini N, Geraci M (2014). A Gradient Search Maximization Algorithm for the Asymmetric Laplace Likelihood, Journal of Statistical Computation and Simulation. http://dx.doi.org/10.1080/00949655.2014.908879. Forthcoming.

lqm

set.seed(123)
n <- 500
test <- data.frame(x = runif(n,0,1))
test$y <- 30 + test$x + rnorm(n)
lqm.ls <- lqm(y ~ x, data = test, fit = FALSE)

do.call("lqm.fit.gs", lqm.ls)