ugomquantreg (version 1.0.0)

ammonia: Ammonia oxidized to acid nitric data set

Description

The data come from experiments with a plant where ammonia is oxidized to acid nitric.

Usage

data(ammonia, package = "ugomquantreg")

Format

A data-frame with 21 observations and 4 columns:

• stackloss: the percentage of ammonia lost.

• airflow: the air flow to the plant.

• watertemp: the cooling water inlet temperature.

• acidconc: the acid concentration.

References

Brownlee, K. A., (1965). Statistical Theory and Methodology in Science and Engineering. New York: John Wiley & Sons.

Yu, K., and Moyeed, R. A., (2001). Bayesian quantile regression. Statistics and Probability Letters, 54(4) 437--447.

Examples

# NOT RUN {
data(ammonia, package = "ugomquantreg")

library(gamlss)

tau <- 0.50
fit.logit <- gamlss(stackloss ~ airflow + watertemp + acidconc, data = ammonia, 
family = UGOM(sigma.link="identity"))

tau <- 0.50
fit.probit <- gamlss(stackloss ~ airflow + watertemp + acidconc, 
data = ammonia, family = UGOM(mu.link = "probit", sigma.link = "log"))

fittaus <- lapply(c(0.10, 0.25, 0.50, 0.75, 0.90), function(Tau){
 tau <<- Tau;
 gamlss(stackloss ~ airflow + watertemp + acidconc, data = ammonia, 
 family = UGOM(mu.link = "logit", sigma.link = "log"))
})

sapply(fittaus, coef)
# }