CODEWITHSUNDEEP
rggplot2distribution
funykcheff
funykcheff

Reputation: 51

ggplot: adding normal distribution curve using stat_function to existing histogram and distribution layers

Task that I need to accomplish: 1. draw x=data/y=density histogram - done 2. draw distribution curve for given dataset - done 3. draw perfect normal distribution curve for this dataset (red line) - problem I assume problem is in 2nd stat_function.

Code is run-able w/o any preparations:

data <- data.frame(c(runif(30,1,50)),c(runif(30,50,1)))
g.data <- data[,1]

graph <- ggplot(data, aes(g.data))
graph <- graph +
geom_histogram(aes(y = ..density..), binwidth = 2, fill = 'pink') +
labs(x = 'Data', y ='Density') +
stat_function(fun = dnorm, args = list(mean = mean(g.data, na.rm = T),
sd = sd(g.data, na.rm =T)), colour ='black', size =1) +
theme(legend.position = 'none') +
stat_function(fun = dnorm, colour = "red", args = list(mean = mean(g.data)))
graph

Here what I get

Here what I approximately need, perfect norm. distribution

Upvotes: 2

Views: 7664

Answers (1)

DeveauP
DeveauP

Reputation: 1237

A normal distribution has 2 parameters: its mean and standard deviation. Here your provide only the mean to dnorm, so it assumes sd = 1.

A corrected version of the code you provide is:

data <- data.frame(c(runif(30,1,50)))
ggplot(data, aes(data[,1])) +
    geom_histogram(aes(y = ..density..), binwidth = 2, fill = 'pink') +
    labs(x = 'Data', y = 'Density') +
    stat_function(fun = dnorm, 
        args = list(mean = mean(data[,1], na.rm = TRUE), 
                    sd = sd(data[,1], na.rm = TRUE)), 
        colour = 'black', size = 1)

enter image description here

Upvotes: 4

Related Questions