← Dashboard
MODULE 4 · ABOUT 5 HOURS

Preparing data

Clean, scale and window time series without peeking at the future.

This module is an outline. The full lecture, widget, quiz and lab are on the way.

Lecture

What you’ll learn

  • Cleaning price data: missing days, splits and adjusted prices
  • Features: returns, rolling volatility, volume
  • Scaling, and why the scaler may only see training data
  • Sliding windows: turning a history into training examples
  • Splitting by date, never at random
  • Walk-forward testing

Lecture outline

  1. Garbage in, garbage out
  2. Building features from prices
  3. Scaling without cheating
  4. Windows: how a sequence model sees time
  5. Train, validation and test, in time order
  6. Data leaks and how to catch them

Discover: Spot the Leak

Five data pipelines, one secretly peeks at the future. Find it.

Interactive widget coming soon.

Quiz

8 to 10 questions. 80% to pass, unlimited retakes.

Quiz coming soon.

Lab: Build make_windows()

A reusable windowing function and a time-ordered train/validation/test split, with built-in checks.

Runs in your browser

Notebook coming soon.