Overlaying a normal pdf onto a histogram in R

Question

I've looked through similar problems and I am unable to resolve mine based on what has been answered with them. I have a histogram of Body Mass Index Data. I have found the mean and sd of the data, and I am trying to overlay a normal pdf with the same mean and sd of the data to compare it with the histogram.

Here is what I have so far.

I used hist(BMI) to graph the histogram. I found the mean(BMI) to be 26.65 and the sd(BMI) to be 3.47.

I am trying to use the curve function to plot a normal curve with those same parameters over the histogram.

curve(dnorm(x, mean = 26.65, sd = 3.47), add = T, col = "red")

HistOverlay

As you can see, the red curve is barely visible at the very bottom of the histogram. Why is this error occurring?

Thank you.

If you want to plot densities, you have to convert histogram to density - divide bins by bin width and be sure integral is equal to 1 — Severin Pappadeux
@SeverinPappadeux Yeah. Or you know just use probability=TRUE in the hist statement. — Dason

CAFEBABE CAFEBABE · Accepted Answer · 2016-02-13T22:16:02

The normal distribution is probability density function. Hence the integral of the function is 1. Simply speaking, it is the probability that a single individual has this BMI. To generate the result you want to have you need to multiply the curve with the size of your population.

Something along the line: curve(100*dnorm(x, mean = 26.65, sd = 3.47), add = T, col = "red")

Overlaying a normal pdf onto a histogram in R

1 Answers