Basic plotting in R

R has very good mechanism to do data plotting.

Let’s explore few of the commands present in R

To plot simple graph-

library(datasets)
head(iris, 300)

plot(iris$Species)

The best part of plot() is that it automatically understands the kind of datatype we are handling and automatically plots the graphs-

In the below example, we are plotting distribution of petal.width-

plot(iris$Petal.Width)

Let’s take one more example of plotting-

plot(iris$Species , iris$Petal.Width)

R will automatically determine that first variable is categorical and second is a numeric and will make the most common plot which is this case is a box plot

The box plot reveals us the quantile-wise distribution of the data.

The box represents Inter Quantile Range IQR(P25 to P75) with dark line being the median.

Lower line is P25–1.5*IQR

Upper line is P75 + 1.5IQR

The dots denote outliers whose values are completely drifting apart from the normal values.

With R , you get different options aswell to decorate the plot in a proper fashion.

Bar Chart

A basic plotting of bar chart can be done via barplot() function

library(datasets)
head(mtcars, 100)

barplot(mtcars$carb)

The above chart displays the values of carburetors in all the models in a graphical representation.

Now, let’s get an aggregated view of carburetors on how many models are having the same value carburetors

library(datasets)
head(mtcars, 100)

?mtcars
barplot(mtcars$carb)

carbFreq <- table(mtcars$carb)
barplot(carbFreq)

The above chart represents number of cars having a given value of carburetors.

table() function takes input as one variable as an input and creates a dataframe which has the frequency of that value occuring in that dataset.

Histogram

Histogram gives us a plot of the frequency of values lying in a range-

library(datasets)

head(iris, 200)

par(mfrow = c(4 , 1))

hist(iris$Sepal.Length)
hist(iris$Sepal.Width)
hist(iris$Petal.Length)
hist(iris$Petal.Width)

The above script gives you different histograms for different columns of iris dataset.