Summary Statistics in Pandas – describe(), agg(), and More in Python 2026

Summary Statistics in Pandas – describe(), agg(), and More in Python 2026

Getting quick and meaningful summary statistics is one of the first steps in any data analysis or manipulation task. In 2026, Pandas provides powerful and flexible ways to compute summary statistics using describe(), agg(), and groupby operations.

TL;DR — Key Methods

• df.describe() – Quick statistical summary for numeric columns
• df.agg() – Custom aggregations with full control
• df.groupby().agg() – Group-wise statistics
• df.value_counts() and df.describe(include="all") for categorical data

1. Basic Summary with describe()

import pandas as pd

df = pd.read_csv("sales_data.csv", parse_dates=["order_date"])

print(df.describe())                    # Numeric columns only
print("
Full summary:")
print(df.describe(include="all"))       # Includes categorical columns

2. Custom Aggregations with agg()

summary = df.agg({
    "amount": ["count", "mean", "median", "min", "max", "std"],
    "quantity": ["sum", "mean"],
    "customer_id": ["nunique"]
}).round(2)

print(summary)

3. Grouped Summary Statistics

grouped = (
    df
    .groupby(["region", "year"])
    .agg({
        "amount": ["sum", "mean", "count"],
        "customer_id": "nunique",
        "order_date": ["min", "max"]
    })
    .round(2)
)

print(grouped)

4. Best Practices in 2026

• Use describe(include="all") to get a quick overview of both numeric and categorical data
• Prefer .agg() over multiple separate calls for better performance
• Combine with groupby() for segmented analysis
• Use .round() to keep output clean
• For very large datasets, consider Dask equivalent methods

Conclusion

Mastering summary statistics with describe() and agg() is essential for exploratory data analysis and reporting. In 2026, using these methods efficiently helps you quickly understand your data, spot outliers, and make informed decisions before diving deeper into manipulation or modeling.

Next steps:

• Run df.describe(include="all") on your next dataset and explore the insights it reveals