Pandas Advanced Aggregations

Introduction

Advanced aggregation techniques using groupby with custom functions.

Custom Aggregations

import pandas as pd

def range_func(x):
    return x.max() - x.min()

def coefficient_of_variation(x):
    return x.std() / x.mean()

df.groupby("category").agg({
    "value": [range_func, "mean", "std"]
})

Named Aggregation (Python 3.7+)

df.groupby("category").agg(
    total="value".sum,
    average="value".mean,
    count=("value", "count")
)

Transform

# Add group statistics back to dataframe
df["group_mean"] = df.groupby("category")["value"].transform("mean")
df["group_normalized"] = df["value"] / df["group_mean"]

Filter Groups

# Keep only groups with certain criteria
df.groupby("category").filter(lambda x: x["value"].sum() > 100)

Practice Problems

1. Create custom aggregation functions
2. Apply multiple transformations
3. Filter groups by criteria
4. Add computed columns to original data
5. Implement rolling aggregations