Uniqueness Property of MGFs
What you'll learn: How the moment generating function (MGF) of a random variable uniquely identifies its entire probability distribution.
The Core Idea
The uniqueness property of MGFs states that if two random variables have the same MGF (and the MGF exists in some interval around zero), then they have the same probability distribution. This is an extraordinarily powerful result: knowing the MGF is equivalent to knowing the entire distribution.
Think of the MGF as a distribution's "fingerprint." Just as no two people share identical fingerprints, no two different distributions share the same MGF. If you can calculate or derive an MGF and recognize it as matching a known distribution's MGF, you've completely identified the distribution—no additional information needed.
Why This Matters
This property is the reason MGFs are so valuable in probability theory. When you:
- Transform random variables and compute the resulting MGF
- Add independent random variables and find the MGF of their sum
- Encounter any complex probability problem that yields an MGF
...you can immediately identify the distribution by matching it to known MGFs, rather than working backwards from moments or probabilities.
For example, if you compute an MGF and find M(t) = e^(λ(e^t - 1)), you instantly know the distribution is Poisson with parameter λ—no need to verify the PMF, mean, variance, or any other property separately.
The Reverse Also Holds
Importantly, different distributions must have different MGFs. If you prove two MGFs differ, you've proven the underlying distributions differ. This makes MGFs useful for showing distributions are distinct.
Key Takeaway: The MGF uniquely determines a probability distribution—if two random variables share the same MGF, they have identical distributions, making MGFs a powerful tool for distribution identification.