Home/

...

/

Numerical variables

Numerical variables

In this lesson, we will learn about numeric variables and their types. We will also cover some graphics to help us visualize this type of variable alongside a few simple lines of code to display them.

Definition

Numerical variables are variables where the value has numerical meaning; for example, age, the number of movies watched, IQ, salary are all represented by numbers.

Numerical variables can be classified as:

  • Continuous variables
  • Discrete variables

Here is an example of age variable as a numerical variable:

Continuous Variable

A variable is continuous if it can assume an infinite number of Real values within a given interval.

As an example, weight (64.2 Kg, 43.8Kg, …), distance (105.7Km, 25.5Km), as you can see, the values are real numbers, that is why it is continuous, temperature and length are all examples of continuous data.

There are various means to visualize continuous variables, to name a few:

  • Box plot
  • Density plot
  • Scatter plot
  • Histogram

Box plot

These previous plot was generated by the following code snippet:

Python 3.5
import seaborn as sns
import numpy as np
import matplotlib.pyplot as plt 
# create simple random continious variables
x = np.random.normal(size=100)
# plot the boxplot
sns.boxplot(x, color = "g")
plt.savefig('output/box.png')

Density plot

The previous density plot was generated by the following code snippet:

Python 3.5
import seaborn as sns
import numpy as np
import matplotlib.pyplot as plt 
# create simple random continious variables
x = np.random.normal(size=100)
# create a density plot of x variable.
sns.distplot(x, hist=False, rug=True)
plt.savefig('output/box.png')

Scatter plot

The previous scatter plot was generated by the following code snippet:

Python 3.5
import seaborn as sns
import numpy as np
import matplotlib.pyplot as plt 
# create simple random continious variables
x = np.random.normal(size=100)
y = np.random.normal(size=100)
# create a scatter plot of x and y variable.
sns.scatterplot(x, y, color = "g")
plt.savefig('output/box.png')

Histograms

The previous histogram plot was generated by the following code snippet:

Python 3.5
import seaborn as sns
import numpy as np
import matplotlib.pyplot as plt 
# create simple random continious variables
x = np.random.normal(size=100)
# creates a histogram plot of x variable with red color.
sns.distplot(x, kde=False, rug=True, color = "r");
plt.savefig('output/box.png')

Discrete variables

A discrete variable cannot take the value of a fraction between one value and the next closest value. It only takes integer values. Examples of discrete variables include the number of registered cars(2, 4, 7), number of business locations(4, 10, 5), and number of children in a family (0, 1, 2), all of which measured as whole units (i.e., 1, 2, 3, …).

Discrete variables can be visualized using:

  • Count plot
  • Pie chart

Count plot

You can use the following code snippet to generate count plots:

Python 3.5
import seaborn as sns
import numpy as np
import matplotlib.pyplot as plt 
# loading the titanic dataset.
titanic = sns.load_dataset("titanic")
# creating a count plot for the class variable.
sns.countplot(x="class", data=titanic)
plt.savefig('output/box.png')

Pie chart

You can always generate a pie chart using the following code snippet:

Python 3.5
import seaborn as sns
import numpy as np
import matplotlib.pyplot as plt 
# loading the titanic dataset.
titanic = sns.load_dataset("titanic")
# getting the count of each class
values = titanic["class"].value_counts().values
# getting the labels of each class.
labels = titanic["class"].value_counts().index
# creating the pie chart.
plt.pie(values, labels= labels,  shadow=True, startangle=90)
plt.savefig('output/box.png')

Access this course and 1200+ top-rated courses and projects.