BALtqr {Brq} | R Documentation |
Bayesian adaptive Lasso tobit quantile regression
Description
This function implements the idea of Bayesian adaptive Lasso tobit quantile regression employing a likelihood function that is based on
the asymmetric Laplace distribution. The asymmetric Laplace error distribution is written as a scale mixture of normal distributions
as in Reed and Yu (2009). The proposed method (BALtqr
) extends the Bayesian Lasso tobit quantile regression by allowing different penalization parameters for different regression
coeffficients (Alhamzawi et al., 2013).
Usage
BALtqr(x,y, tau = 0.5, left = 0, runs = 11000, burn = 1000, thin=1)
Arguments
x |
|
y |
|
tau |
|
left |
|
runs |
|
burn |
|
thin |
|
Author(s)
Rahim Alhamzawi
References
[1] Alhamzawi, Rahim. (2013). Tobit Quantile Regression with the adaptive Lasso penalty. The Fourth International Arab Conference of Statistics, 450 ISSN (1681 6870).
[2] Reed, C. and Yu, K. (2009). A partially collapsed Gibbs sampler for Bayesian quantile regression. Technical Report. Department of Mathematical Sciences, Brunel University. URL: http://bura.brunel.ac.uk/bitstream/2438/3593/1/fulltext.pdf.
Examples
# Example
n <- 150
p=8
Beta=c(5, 0, 0, 0, 0, 0, 0, 0)
x <- matrix(rnorm(n=p*n),n)
x=scale(x)
y <-x%*%Beta+rnorm(n)
y=y-mean(y)
y=pmax(0, y)
fit = Brq(y~0+x,tau=0.5, method="BALtqr",runs=5000, burn=1000)
summary(fit)
model(fit)