Skip to content

Lesson 5 of 6 · 12 min

Validation

Check a simulation against held-out observations.

Validate structure (does the model behave plausibly?) and outputs (does it reproduce held-out data?). Report the error metric next to every result.

A model can be validated for one question and invalid for another. State the validity range.

Key points

  • Validate against data the model did not see

Knowledge check

Uncertainty

Monte Carlo

Why is a single simulation run insufficient for a policy decision?

Choose one answer
Answers are checked anonymously unless you sign in.

Sign in to mark this lesson complete and resume where you left off.