On this page
What this lesson covers
Multivariate Monte Carlo simulation generates several random inputs jointly so their simulated dependencies match the specified model.
The lesson builds correlated normal variables from independent standard normal draws. It uses a two-variable Cholesky transformation to preserve standard deviations and correlation, then evaluates one scenario.
How do you simulate two correlated normal variables?
Scale one independent standard normal draw for the first zero-mean variable, then combine both draws with correlation weights before scaling the second.
First zero-mean normal variable X1 = sigma1 × Z1
- First variable standard deviation
- First independent standard normal draw
Second zero-mean normal variable X2 = sigma2 × (rho × Z1 + sqrt(1 − rho²) × Z2)
- Second variable standard deviation
- Target correlation
- Independent standard normal draws
For example, assume jointly normal, zero-mean factors with share-return volatility 0.9%, yield-change volatility 5 bp and correlation 0.28. With independent standard normal draws Z1 = −2 and Z2 = 0.5, the share return is 0.9% × (−2) = −1.8%; the yield change is 5 × [0.28 × (−2) + 0.96 × 0.5] = −0.4 bp.
Common mistakes
- Drawing dependent risk factors separately removes their relationship from the scenarios.
- Expecting correlation to force matching directions in every scenario mistakes a tendency across scenarios for a rule governing each one.
Try a practice question
Choose an answer, then check it to read the explanation for every option.
Continue studying in Quizara
In Quizara the lesson comes with more practice questions on the topic, explanations for every option and an AI tutor for your follow-up questions.
Written by the Quizara Editorial Team under our editorial standards.