Can't add a probability-curve on the histogram

Question

I'm trying do display multiple histograms with one plot with the lattice-package.

That's my code so far:

histogram(~ X1 + X2 + X3 + X4 + X5 + X6 + X7 + X8 + X9 + X10, data=mydata, 
      type = "density",layout=c(5,2),
      panel=function(x, ...) {
        panel.histogram(x, ...)
        panel.mathdensity(dmath=dnorm, col="black",
                          args=list(mean=mean(x), sd=sd(x)), ...)
      })

The problem is, that it won't plot the probability-curve. It doesn't give me an error back, so the code looks good, I think.

I also tried it with only one variable and it didn't work either:

histogram(~ X1, data=mydata, 
  type = "density",layout=c(5,2),
  panel=function(x, ...) {
    panel.histogram(x, ...)
    panel.mathdensity(dmath=dnorm, col="black",
                      args=list(mean=mean(x), sd=sd(x)), ...)
  })

Does anyone see an error in my code? Or could be something wrong in my data?

I'm glad for any advice!

We can really only help you if you make your example code reproducible. Otherwise we have to guess at the structure of your data. See http://stackoverflow.com/q/5963269/602276 how to do this. — Andrie, Dec 05 '12 at 13:07

score 4 · Answer 1 · answered Dec 16 '12 at 11:29

Could it be that your data contain missing values?

# Create example data (no missings)
mydata <- data.frame(X1 = rpois(1000, 12), X2 = rnorm(1000, 12, sqrt(12)))

# Create some missing (NA) entries
mydata2 <- mydata
mydata2[sample(seq_len(nrow(mydata2)), 10), 1] <- NA

Using the above mydata2 object in the histogram function produces no density plot for X1, since mean and sd return NA. Adding na.rm = TRUE to both those functions will return values that panel.mathdensity can use:

histogram(~ X1 + X2, data=mydata2, 
      type = "density",layout=c(1,2),
      panel=function(x, ...) {
        panel.histogram(x, ...)
        panel.mathdensity(dmath=dnorm, col="black",
# Add na.rm = TRUE to mean() and sd()
                          args=list(mean=mean(x, na.rm = TRUE),
                                    sd=sd(x, na.rm = TRUE)), ...)
      })

histogram with density

score 0 · Answer 2 · answered Dec 05 '12 at 13:16

Without your data it is hard to help you.

This is a simple example, maybe can help you. I try to keep your settings and correct some ones.

library(lattice)

dat <- data.frame(X1 = rnorm(10000),Y1 =rnorm(10000))
histogram(~X1+Y1,
          data = dat,
          main=list(
            label="Main plot title",
            cex=1.5),
          xlab=list(
            label="Custom x-axis label",
            cex=0.75),
          ylab=list(
            label="Your Y label ",
            cex=1.2),
          scales=list(cex=0.5),
          layout = c(1,2),
          par.settings = list(
                              type = "density",
                              panel=function(x, ...) {
                                panel.histogram(x, ...)
                                panel.mathdensity(dmath=dnorm, col="black",
                                                  args=list(mean=mean(x), sd=sd(x)), ...)
                              })
)

enter image description here

Thanks, but i didn't work. I can't see a probability curve on the image you posted? — Ventrue12, Dec 06 '12 at 07:03

Can't add a probability-curve on the histogram

2 Answers2

Linked