Quant Sandbox
DerivativesPortfolio & RiskFixed IncomeCorporate FinanceProbability & MeasureStochastic ModelsMarket MicrostructureGuide
Stochastic Models
Random Walk
Random Walk labSymmetric Random WalkRescaled Random WalkStandard deviationCentral Limit Theorem
Brownian Motion
Brownian labDefining the Wiener processIncrement explorerJoint distributions and covarianceBrownian transformations and invarianceQuadratic variationBrownian bridgeRunning maximum and minimumReflection principleFirst passage timesHitting probabilities
Geometric Brownian Motion
LabDefinition: ABM vs GBMEuler–MaruyamaMean reversionJumpsCorrelated motions
Itô Process
LabItô integralItô's lemmaMartingalesGirsanov
Monte Carlo
LabVariance reductionFeynman–Kac
Connected Guides
  • Geometric Brownian motion
  • Returns and volatility
  • Change of measure and Girsanov's theorem
  • Monte Carlo option pricing
  • Simulating SDEs: the Euler–Maruyama scheme

© 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.

Geometric Brownian motion, defined against arithmetic Brownian motion

Both models below start at S₀ = 100 and are driven by exactly the same random shocks. The only difference is how a shock is applied: ABM adds a fixed amount, GBM multiplies by a percentage.

Parameters

8%
-0.30.3
35%
0.050.8
3.0
0.510
Random seed: 7
Metrics
ABM: P(S_T < 0)
2.0%
ABM paths below 0 at T
2.3%
ABM st. dev. at T
60.6
GBM st. dev. at T
84.7
GBM median vs mean
105.8 / 127.1
The two definitions side by side
Arithmetic BMGeometric BM
DynamicsdSt=μS0 dt+σS0 dWtdS_t = \mu S_0\,dt + \sigma S_0\,dW_tdSt​=μS0​dt+σS0​dWt​dSt=μSt dt+σSt dWtdS_t = \mu S_t\,dt + \sigma S_t\,dW_tdSt​=μSt​dt+σSt​dWt​
SolutionSt=S0+μS0t+σS0WtS_t = S_0 + \mu S_0 t + \sigma S_0 W_tSt​=S0​+μS0​t+σS0​Wt​St=S0 e(μ−12σ2)t+σWtS_t = S_0\, e^{(\mu - \frac{1}{2}\sigma^2)t + \sigma W_t}St​=S0​e(μ−21​σ2)t+σWt​
A shock moves the price bya fixed amount σS₀√dt·Z, the same at every price levela percentage σ√dt·Z of the current price, so moves grow with the price
S_T isnormal: symmetric around its mean, unbounded on both sideslog-normal: skewed to the right, the log of S_T is normal
Can go below 0?Yes, with positive probabilityNever: e^x > 0
σ is measured inprice units (here σS₀ per √year)returns: percent per √year
Typical usespreads, P&L, the Bachelier model, rates that may turn negativestock prices and indices, the Black–Scholes model
Same shocks, two models (400 paths)
Arithmetic BM
Geometric BM

The highlighted path is the same sequence of shocks in both charts. ABM paths fan out evenly and can cross zero (red line); GBM paths fan out more on the upside and squeeze together near zero without ever reaching it.

Distribution of S_T at T = 3
Arithmetic BM
Geometric BM

Bars: simulated end values. Curves: exact laws, normal for ABM and log-normal for GBM. Both share the same axis.

Step size against price level
Arithmetic BM
Arithmetic BM
Geometric BM
Geometric BM

Each dot is one step: the price before it (x) and how far it moved (y). ABM steps form a flat band, the same size at any price. GBM steps form a fan that widens with the price, because the shock is a percentage.