spBQR: Spatial Bayesian quantile regression models

Description Usage Arguments Value References Examples

View source: R/spBQR.R

Description

This function estimates a spatial Bayesian quantile regression model The Asymmetric Laplace Process (ALP) is considered to fit this spatial model.

Usage

1
2
3
4
spBQR(formula, tau = 0.5, data, itNum, thin = 1, betaValue = NULL,
  sigmaValue = 1, spCoord1, spCoord2, lambda = 0.5, tuneP = 1,
  alpha = 0.5, tuneA = 1000, priorVar = 100, refresh = 100, quiet = T,
  jitter = 0, includeAlpha = TRUE, tuneV = 0.5, kMT = 5)

Arguments

formula

a formula object, with the response on the left of a ~ operator, and the terms, separated by + operators, on the right.

tau

Quantile of interest.

data

A data.frame from which to find the variables defined in the formula

itNum

Number of iterations.

thin

Thinning parameter. Default value is 1.

betaValue

Initial values for the parameter beta for the continuous part.

sigmaValue

Initial value for the scale parameter.

spCoord1

Name of the first spatial coordinate, as character.

spCoord2

Name of the second spatial coordinate, as character.

lambda

Initial value for the parameter in the covariance matrix.

tuneP

Tuning parameter for the Metropolis-Hastings algorithm to draw samples from the posterior distribution of kappa.

alpha

Value between 0 and 1 of the pure error variance in the covariance matrix. Default is 0.5.

tuneA

Tuning parameter for the Metropolis_Hastings algorithm to draw samples from the posterior distribution of alpha.

priorVar

Value that multiplies an identity matrix in the elicition process of the prior variance of the regression parameters.

refresh

Interval between printing a message during the iteration process. Default is set to 100.

quiet

If TRUE, the default, it does not print messages to check if the MCMC is actually updating. If FALSE, it will use the value of refresh to print messages to control the iteration process.

jitter

add a small value to invert spatial correlation matrix

includeAlpha

If TRUE, the default, the model will include the alpha parameter. If FALSE, alpha is set to zero for all draws of the chain.

tuneV

Tuning parameter to the multiple-try Metropolis to sample for the posterior distribution of the latent variables. Default value is 0.5.

kMT

Integer, number of Metropolis samples in the multiple-try Metropolis. Default value is 5.

Value

A list with the chains of all parameters of interest.

References

Lum and Gelfand (2012) - Spatial Quantile Multiple Regression Using the Asymmetric Laplace process. Bayesian Analysis.

Examples

1

brsantos/baquantreg documentation built on Dec. 10, 2018, 1 p.m.