barplot in loop in R

Question

Here is my question:

#data 1:
lab1 <- 1:10
group <- rep(1:3, each = length (lab1))
label <- rep(lab1, 3)
avar <-  rep(c(0, 1, 4, 5, 6, 8, 10,  11, 12,  13), 3)
myd <- data.frame (group, label, avar)

# data 2
fillcol <- rep(rnorm(length(lab1)-1, 0.5, 0.2), 3)
group1 <- rep(1:3, each = length(fillcol)/3)
 # this variable will be used to fill color in bars
 filld <- data.frame(group1, fillcol)

# now plotting
par(mfrow = c(3, 1))
 par(mar = c(2.5, 1, 2.5, 1))

#plot1
myd1 <- myd[myd$group ==1,]
filld1 <- filld[filld$group1 ==1,]
blues <- colorRampPalette(c("yellow", "blue"))
barplot(as.matrix(diff(myd1$avar)), horiz=T, col=blues(10)[10* filld1$fillcol], 
 axes=F, xlab="Mark")
axis(1, labels=myd$label, at=myd$avar)
axis(3, labels=myd$avar, at=myd$avar)

Although this sample dataset, I have many variables and I want to automate this process.

for (i in 1:length(unique(myd$group))){
         par(mfrow = c(i, 1))
        par(mar = c(2.5, 1, 2.5, 1))
                myd[i] <- myd[myd$group ==i,]
        filld[i] <- filld[filld$group1 ==i,]
      blues <- colorRampPalette(c("yellow", "blue"))
      barplot(as.matrix(diff(myd[i]$avar)), horiz=T, 
   col=blues(10)[10* filld1$fillcol], axes=F, xlab="Mark")
      axis(1, labels=myd[i]$label, at=myd[i]$avar)
        axis(3, labels=myd[i]$avar, at=myd[i]$avar)
        }
 Error in dim(data) <- dim : attempt to set an attribute on NULL
In addition: Warning messages:
1: In `[<-.data.frame`(`*tmp*`, i, value = list(group = c(1L, 1L, 1L,  :
  provided 3 variables to replace 1 variables
2: In `[<-.data.frame`(`*tmp*`, i, value = list(group1 = c(1L, 1L,  :
  provided 2 variables to replace 1 variables

Edits: I want to create a loop so that I create multiple graphs within the loop:

PS: I am new to R and stackoverflow. Please excuse me if the question is not appropriate, although I read the guidelines and try to adhere to it

Answer 1

If you wish to produce series of barplots then your code seems fine to me. If you wish to produce a bar chart with the subgroups then you can take a look at this link: http://www.harding.edu/fmccown/r/

Answer 2

If you wrap it in a function like this, you should be able to get it to work and be called multiple times.

colbarplot <- function(group) {

    myd1 <- myd[myd$group == group,]
    filld1 <- filld[filld$group1 == group,]
    blues <- colorRampPalette(c("yellow", "blue"))
    barplot(as.matrix(diff(myd1$avar)), horiz=T,
            col=blues(10)[10* filld1$fillcol], 
            axes=F, xlab="Mark")
    axis(1, labels=myd$label, at=myd$avar)
    axis(3, labels=myd$avar, at=myd$avar)
}

par(mfrow = c(3, 1))
par(mar = c(2.5, 1, 2.5, 1))
sapply(unique(myd$group),function(x) colbarplot(x))

This will give the multiple plots on one page you are after.

Answer 3

Two problems here. You're unintentionally overwriting 'myd' and 'filld' in the loop. Second, you should specify the mfrow at the start, then produce the plots (3 in this example). You're resetting mfrow in the middle. Not what you want.

Try this:

opar <- par(mfrow = c(length(unique(myd$group)), 1), mar = c(2.5, 1, 2.5, 1))

for (i in 1:length(unique(myd$group))){
    myd1 <- myd[myd$group ==i,]
    filld1 <- filld[filld$group1 ==i,]
    blues <- colorRampPalette(c("yellow", "blue"))
    barplot(as.matrix(diff(myd1$avar)), horiz=T, 
       col=blues(10)[10* filld1$fillcol], axes=F, xlab="Mark")
    axis(1, labels=myd1$label, at=myd1$avar)
    axis(3, labels=myd1$avar, at=myd1$avar)
 }

 par(opar);