如何在R中将X轴拆分为十分位数并进行ggplot

Question

Hi I am wondering how to split x-axis as decile in R and make ggplot?

I currently have age range data and NO2 pollution data. The two datasets share the same geographic reference named ward. I wish to plot my demographic data in quantiles of equal number of ward (Total 298).

I tried the quantile regression in R where I used the following:

library(SparseM) 
library(quantreg)
mydata<- read.csv("M:/Desktop10/Test2.csv") 
attach(mydata) 
Y <- cbind(NO2.value)
X <- cbind(age.0.to.4, age..5.to.9, age.10.to.14, age.15.to.19, age.20.to.24, age.25.to.29, age.30.to.44, age.45.to.59, age.60.to.64, age.65.to.74, age.75.to.84, age.85.to.89, age.above.90) 
quantreg.all <- rq(Y ~ X, tau = seq(0.05, 0.95, by = 0.05), data=mydata) 
quantreg.plot <- summary(quantreg.all) 
plot(quantreg.plot) 

But what I get are not what I expected as the y-axies is not the NO2 data.

The ideal plot is attached:

Many thanks for your help and suggestions.

Answer 1

If I understand your question, I think the cut function combined with the quantile function will create the deciles. Here's an example with fake data.

In the code below, we use the cut function to split the data into deciles and we use the quantile function to set the breaks argument for cut . This tells cut to group the data into 10 groups of equal size, from smallest values of NO2 to largest.

group_by(age) means we create the deciles separately for each age group. This means that there are equal numbers of subjects within each decile in a given age group, but the NO2 cutoff values for each decile are different for different age groups. To create deciles over the data as a whole, just remove group_by(age) . This will result in the same NO2 cutoff values for each decile across all age groups, but within a given age group, the number of subjects will not be the same in each decile.

library(tidyverse)

# Fake data
set.seed(2)
dat = data.frame(NO2=c(runif(600, 0, 10), runif(400, 1, 11)), 
                 age=rep(c("0-10","11-20"), c(600,400)))

# Create decile groups
dat = dat %>% 
  group_by(age) %>% 
  mutate(decile = cut(NO2, breaks=quantile(NO2, probs=seq(0,1,0.1)), 
                      labels=10:1, include.lowest=TRUE),
         decile = fct_rev(decile))

Now we plot using ggplot2 . The stat_summary function returns the mean for each decile in each age group.

ggplot(dat, aes(decile, NO2, colour=age, group=age)) +
  stat_summary(fun.y=mean, geom="line") +
  stat_summary(fun.y=mean, geom="point") +
  expand_limits(y=0) +
  theme_bw()