Moment Generating Functions

Probability · Axiom Academy

LESSON Moment Generating Functions Use exponential transformations to extract moments and characterize distributions uniquely For a random variable X (discrete or continuous), the moment generating function is defined as: The name comes from the fact that derivatives of M(t) at t = 0 generate the moments of X. 2. Extracting Moments via Derivatives The k-th moment E[X^k] is obtained by taking the k-th derivative of M(t) and evaluating at t = 0: This follows from differentiating under the expectation. Each derivative brings down a factor of X, so M^(k)(0) = E[X^k]. One of the most important properties of MGFs is their uniqueness: This means the MGF completely determines the distribution. If you can identify the form of M(t), you've identified the distribution of X. Like PGFs, MGFs have a multiplication property for independent sums: Since E[e^ t(X+Y) ] = E[e^ tX e^ tY ] and independence allows factoring expectations, the MGF of a sum is the product of individual MGFs. This is crucial for the Central Limit Theorem. 5. Example: Normal Distribution The normal distribution N(μ, σ²) has the MGF: From this, we can verify E[X] = M'(0) = μ and Var(X) = M''(0) - [M'(0)]² = σ². The sum of independent normals is normal follows from multiplying MGFs.

This is the written version of the interactive lesson above. See the full Probability course.