Loop by ID with ggplot2

Question

I have a dataset looking like this:

#ID   CWRES
#1    0.90000
#1    0.59034
#1    1.81300
#1    1.42920
#1    0.59194
#2    1.90000
#2    1.09034
#2    2.01300
#2    0.42920
#2    0.19194
#...
#40   1.90000
#40   1.59034
#40   0.81300
#40   0.42920
#40   0.59194

That is one variable (CWRES) and 40 individuals (ID) I want to plot an histogram of every ID in a loop using ggplot. I know how to do it with a "normal" syntax, but I am not able to translate this syntax in ggplot.

Here is the "normal" syntax:

#attach(tab)
#for(i in unique(ID)) { 
#    j = ID==i
#    hist(CWRES,probability = T, ylab="Frequency",xlab=paste(CWRES_label),
#         col="grey90",cex=1,main=paste(paste("ID: #",i,sep="")),nclass=20)
#    lines(
#      density(CWRES,col="blue",lwd=2,lty=1))
#    box()
#    abline(v=mean(CWRES),col="blue",lwd=2)
#    abline(v=quantile(CWRES,0.025),col="blue",lwd=2,lty=2)
#    abline(v=quantile(CWRES,0.50),col="blue",lwd=2,lty=2)
#    abline(v=quantile(CWRES,0.975),col="blue",lwd=2,lty=2)
#}
#detach()

Does anybody know how to translate this in to ggplot syntax?

Thank you very much in advance,

Best regards,

Mario

You mean you want to create 40 separate histograms (for example, one on each page of a PDF?) — David Robinson, Sep 16 '14 at 00:37
By the way, the code you provide does not work as you intend it to. You never use the variable `j`, which means that every histogram is exactly the same (using all of CWRES, rather than the subset for that ID). — David Robinson, Sep 16 '14 at 00:38
[This question and answers](http://stackoverflow.com/questions/19146665/how-subset-a-data-frame-by-a-factor-and-repeat-a-plot-for-each-subset/19147917?noredirect=1#comment39128285_19147917) may be useful to you (plots by group using lists). — aosmith, Sep 16 '14 at 16:54
You are right David, I should not use j in the loop. Thanks to point this out and for your suggestion. I really appreciate it :) — Mario, Sep 16 '14 at 23:13

Paulo E. Cardoso · Accepted Answer · 2014-09-16T01:27:44.680

Would something like this help?

library(ggplot2)
for(i in unique(df$ID)){
  gi <- ggplot(aes(x = CWRES), data = subset(df, ID == i)) +
    geom_histogram()
  ggsave(filename = sprintf('%s.png', i), plot = gi)
}

or wrap it to a function

f.Plotit <- function(varID) {
  subdf = subset(df, ID == varID)
  g1 = ggplot(data = subdf, aes(x = CWRES)) +
    geom_histogram()
  print(g1)
  ggsave(sprintf("%s.png", varID))
}
lapply(unique(df$ID), f.Plotit)

Don't forget to check your working dir getwd() or change it accordingly...

Loop by ID with ggplot2

1 Answers1