pythonintermediate

Pandas Time Series Analysis

Time series operations with resampling, rolling windows, date offsets, and period conversions.

pythonPress ⌘/Ctrl + Shift + C to copy

import pandas as pd
import numpy as np

# Create time series data
dates = pd.date_range("2024-01-01", periods=365, freq="D")
df = pd.DataFrame({
    "date": dates,
    "sales": np.random.randint(100, 1000, 365).astype(float),
    "visitors": np.random.randint(500, 5000, 365),
})
df = df.set_index("date")

# Resample to weekly/monthly aggregates
weekly = df.resample("W").agg({"sales": "sum", "visitors": "mean"})
monthly = df.resample("ME").agg({"sales": "sum", "visitors": "sum"})

# Rolling window calculations
df["sales_7d_avg"] = df["sales"].rolling(window=7).mean()
df["sales_30d_avg"] = df["sales"].rolling(window=30).mean()
df["sales_7d_std"] = df["sales"].rolling(window=7).std()

# Expanding window (cumulative)
df["cumulative_sales"] = df["sales"].expanding().sum()

# Percent change
df["daily_change"] = df["sales"].pct_change()
df["weekly_change"] = df["sales"].pct_change(periods=7)

# Shift for lag features
df["sales_yesterday"] = df["sales"].shift(1)
df["sales_last_week"] = df["sales"].shift(7)

# Filter by date range
q1 = df.loc["2024-01":"2024-03"]
jan = df.loc["2024-01"]

print(f"Monthly summary:\n{monthly.head()}")
print(f"\nQ1 total sales: {q1['sales'].sum():,.0f}")