Skip to content

Quantitative Methods & Risk·Monte Carlo & Distributions

Monte Carlo Intuition

11 min read

Run the experiment a thousand times

Monte Carlo simulation is the technique of running the same probabilistic experiment thousands of times to estimate the distribution of outcomes. In trading, you specify a process for asset prices (drift, vol, possibly jumps or regime switches), simulate many paths forward, and observe the distribution of terminal P&L. The technique was developed at Los Alamos in the 1940s — every modern risk system descends from it.

$100 portfolio, 60-day forward sim80 paths · 60 steps · μ=0.0004 σ=0.012

Terminal distribution

60 daily steps, μ=0.04% drift, σ=1.2% daily vol. The terminal-distribution histogram below shows where most outcomes cluster.
Tip

Why not analytical?

For simple processes (constant drift + vol, single asset), closed-form solutions exist. The moment you add path-dependence, optionality, regime switches, or correlated multi-asset portfolios, Monte Carlo is the only way to get a tractable answer.