Quant Sandbox
DerivativesPortfolio & RiskFixed IncomeCorporate FinanceStochastic ModelsGuide
Stochastic Models
Stochastic Learning StudioBeta
1.1Symmetric Random Walk1.2Rescaled Random Walk1.3Standard deviation1.4Central Limit Theorem2.1Defining the Wiener process2.2Increment explorer2.3Joint distributions and covariance3.1Running maximum and minimum3.2Reflection principle3.3First passage times3.4Hitting probabilities
Brownian MotionBeta
Geometric Brownian MotionBeta
Itô ProcessBeta
Monte CarloBeta
Curriculum/Chapter 3 · Brownian Path Geometry/3.1
← PreviousJoint distributions and covarianceNext →Reflection principle

© 2026 Quant Sandbox · A study aid for quantitative finance

Privacy & dataNo cookies

This site does not use cookies and asks for no personal data. It keeps a few small settings in your browser's local storage — your chosen theme and language, the sidebar layout, and your progress and answers in the Guides — so it remembers them on your next visit. This data never leaves your device; clearing your browser storage removes it. Inputs you change can also be captured in the page URL, so you can share or bookmark a specific setup. The site uses Vercel Web Analytics, a privacy-friendly, cookieless service that records only anonymous, aggregated usage (such as page views) without identifying you or tracking you across other sites.

3.1

Running maximum and minimum

How high does a Brownian path climb by a given time, and how is that peak distributed?

Goal: follow the running maximum and minimum through time and compare the law of M(t) with |W(t)| and the density 2φ(m/√t)/√t.

Simulation controls

Observation timeDrag or play: the maximum can only rise, the minimum can only fall
1.00
0.001.00
1,000
1003000
7
19999
Predict first

What is the average running maximum of the path by t = 1?

Empirical vs theory

E[M(t)] empirical
?
√(2t/π)
?
E[M(t) − m(t)] empirical
?
2√(2t/π)
?

The maximum measured on the grid always slightly underestimates the true one: the path can climb higher between grid points. The gap is of order √Δt.

Try this

  1. Predict first: what is the average running maximum at t = 1? Check it against the E[M(t)] card.
  2. Play time forward: does the green running maximum ever step down? Why not?
  3. Quadruple t (0.25 → 1): by what factor does E[M(t)] grow? The same √t law as in lesson 1.3.
Brownian path with its running maximum and minimum
t = 1.00
  • Highlighted path
  • Running max M(t)
  • Running min m(t)

The running maximum climbs in steps: it only increases when the path sets a new high. The faint cloud shows the other simulated paths up to the same time.

Distribution of M(t)
M(t) empirical|W(t)| empirical2φ(m/√t)/√t

The empirical densities of M(t) and |W(t)| both match the theoretical 2φ(m/√t)/√t — two different random variables with the same distribution.