R programming | Kernel density plots
Kernel density estimation is a nonparametric method for estimating the probability density function of a random variable. Basically, we’re trying to draw a smoothed histogram, where the area under the curve equals 1. Kernel density plots can be an effective way to view the distribution of a continuous variable. The format for a density plot is:
ggplot(data, aes(x = contvar)) + geom_density()
where data is a data frame and contvar is a continuous variable
The density plot can be filled with color.
Smoothness of the curve is controled with a bandwidth parameter. Using a larger bandwidth will give a smoother curve with fewer details. A smaller value will give a more jagged curve.
bw.nrd0(cars2008$cty) # display the current bandwidth (1.41)
ggplot(cars2008, aes(x=cty)) +
geom_density(fill="red", bw=.5) +
labs(title="Kernel density plot with bw=0.5",
x="City Miles Per Gallon")
Kernel density plots can be used to compare groups. For this example, we’ll compare the 2008 city gas mileage estimates for four-, six-, and eight-cylinder cars using both colored and filled density plots:
ggplot(cars2008, aes(x=cty, color=Cylinders, linetype=Cylinders)) + geom_density(size=1) +
labs(title="Fuel Efficiecy by Number of Cylinders", x = "City Miles per Gallon")
ggplot(cars2008, aes(x=cty, fill=Cylinders)) +
geom_density(alpha=.4) +
labs(title="Fuel Efficiecy by Number of Cylinders", x = "City Miles per Gallon")
#rprogramming
#rstudio
#ggplot
#kerneldensity
#kernel
#kerneldensitycurve
#rdatacode