Box plot – Python(matplot)

In Matplotlib, Box Plot helps visualize data distribution by intuitively showing the median, quartiles, outliers, and more. It often plays a key role in statistical analysis for comparing data distribution and variability.

• Median (Median): The 50th percentile of the data (Q2).
• Quartiles:
  • Q1 (1st quartile): The 25th percentile of the data.
  • Q3 (3rd quartile): The 75th percentile of the data.
• IQR (Interquartile Range): Q3 – Q1.
• Whiskers:
  • Data that typically falls within the range of Q1 – 1.5×IQR and Q3 + 1.5×IQR.
  • Values outside of these ranges are labeled as Outliers.
• Outliers: Points outside the whisker range.

Basic Usage: plt.boxplot()

The default function to plot is plt.boxplot().

plt.boxplot(
    x,                  # data (list or array)
    labels=None,        # labels for each box
    notch=False,        # show median confidence interval (True/False)
    vert=True,          # vertical boxplot (False: horizontal)
    patch_artist=False, # set box style (True: color fill)
    showmeans=False,    # show mean values (True)
    meanline=False,     # show means as a line (True)
    showfliers=True,    # show outliers (True/False)
    whis=1.5            # set whisker range (default: 1.5×IQR)
)

Box Plot Code

import matplotlib.pyplot as plt
import numpy as np

np.random.seed(10)
data = [np.random.normal(0, std, 100) for std in range(1, 5)]

plt.boxplot(data, vert=True, patch_artist=True)
"""
notch=True : Shows a notch to show the confidence interval of the median.
vert=False : Displays the box horizontally.
patch_artist=True : Applies a style to fill the box.
"""
plt.xlabel('X-axis')
plt.ylabel('Y-axis')
plt.title('Box Plot Example')
plt.show()

Benefits

• Compare distributions: Compare distributions across multiple data sets at a glance.
• Identify outliers: Easily identify outliers.
• Simplicity: Summarize complex data in a concise way.

Cons

• Lack of detail: Lack of detailed distribution information compared to histograms or violin plots.
• Potential for misinterpretation: Different medians and means can lead to interpretation errors.

Box Plot vs Violin Plot

Organizing Box Plot Information

• Effective statistical visualization: They help present key statistical information in a clear and structured way.
• Intuitive visualization: Box plots clearly show the distribution and outliers in your data.
• Functionality: The plt.boxplot() function creates box plots and supports customization with parameters like notch, patch_artist, and whis.
• Customization: You can enhance readability by adjusting colors, line styles, and labels.
• Comparison & analysis: Box plots are useful for comparing data across different groups and displaying statistical summaries.