Downsample and Aggregate Time Series with resample() in Pandas

Question

Given a high-frequency transaction log indexed by Datetime, downsample the data into calendar month-end buckets ('ME') and compute total monthly revenue and transaction count.

Accepted Answer

Optimal Solution:
import pandas as pd

# Set timestamp as DatetimeIndex
df['timestamp'] = pd.to_datetime(df['timestamp'])
df = df.set_index('timestamp')

# Resample to Month End ('ME' in pandas 2.2+)
monthly_df = df.resample('ME').agg(
    total_revenue=('amount', 'sum'),
    txn_count=('amount', 'count')
).reset_index()

Technical Explanation:
`.resample('ME')` creates time-based bins (like SQL `DATE_TRUNC`) on a DatetimeIndex and applies custom aggregations.

timestamp	amount
2026-01-05 10:00:00	500
2026-01-20 14:00:00	750
2026-02-02 09:00:00	1200

timestamp	total_revenue	txn_count
2026-01-31 00:00:00	1250	2
2026-02-28 00:00:00	1200	1