frame_br: Visualization of quantile regression model fitting: br...

In quokar: Quantile Regression Outlier Diagnostics with K Left Out Analysis

Description

get the observation used in br algorithem

Usage

frame_br(object, tau)

Arguments

object	quantile regression model using br method
tau	quantiles can be a single quantile or a vector of quantiles

Details

This is a function that can be used to create point plot for the observations used in quantile regression fitting based on 'br'method.

Value

All observations and the observations used in quantile regression fitting using br algorithem

Author(s)

Wenjing Wang [e-mail]

Examples

library(ggplot2)
library(quantreg)
data(ais)
tau <- c(0.1, 0.5, 0.9)
object1 <- rq(BMI ~ LBM, tau, method = 'br', data = ais)
data_plot <- frame_br(object1, tau)$all_observation
choose <- frame_br(object1, tau)$fitting_point
ggplot(data_plot,
 aes(x=value, y=data_plot[,2])) +
 geom_point(alpha = 0.1) +
 ylab('y') +
 xlab('x') +
 facet_wrap(~variable, scales = "free_x", ncol = 2) +
 geom_point(data = choose, aes(x = x, y = y,
                                      group = tau_flag,
                                      colour = tau_flag,
                                      shape = obs))

object2 <- rq(BMI ~ Ht + LBM + Wt, tau, method = 'br',
            data = ais)
data_plot <- frame_br(object2, tau)$all_observation
choose <- frame_br(object2, tau)$fitting_point
ggplot(data_plot,
 aes(x=value, y=data_plot[,2])) +
 geom_point(alpha = 0.1) +
 ylab('y') +
 xlab('x') +
 facet_wrap(~variable, scales = "free_x", ncol = 2) +
 geom_point(data = choose, aes(x = x, y = y,
                                      group = tau_flag,
                                      colour = tau_flag,
                                      shape = obs))

quokar documentation built on May 2, 2019, 6:39 a.m.