Independence of More Than Two Random Variables
What you'll learn: How to extend the concept of independence from pairs of random variables to any finite collection of random variables.
The Core Idea
When working with two random variables, we say they're independent if their joint distribution factors into the product of their marginals. But what happens when you have three, four, or n random variables?
For a collection of random variables to be mutually independent, the joint distribution must factor completely into the product of all individual marginal distributions. This is stronger than just checking independence pairwise.
Mutual Independence Defined
Random variables X₁, X₂, ..., Xₙ are mutually independent if and only if:
Discrete case:
P(X₁ = x₁, X₂ = x₂, ..., Xₙ = xₙ) = P(X₁ = x₁) · P(X₂ = x₂) · ... · P(Xₙ = xₙ)
Continuous case:
f(x₁, x₂, ..., xₙ) = f₁(x₁) · f₂(x₂) · ... · fₙ(xₙ)
for all possible values.
Why "Mutual" Matters
Here's the subtlety: pairwise independence does not imply mutual independence. You could have three variables where any two are independent, but all three together are not. It's like three friends where any two get along fine, but putting all three in a room creates conflict—the trio dynamic is different from the pair dynamics.
For true mutual independence, every subset must also factor correctly. For three variables X, Y, Z, you need the joint of all three to factor, and you need X and Y independent, X and Z independent, and Y and Z independent.
Key Takeaway: Mutual independence of n random variables requires complete factorization of the joint distribution into all n marginals—pairwise independence alone is insufficient.