Functions of Random Variables: Discrete Case
You'll learn how to find the probability distribution of a transformed discrete random variable.
When you apply a function to a discrete random variable X, you create a new random variable Y = g(X). The key question is: what's the probability mass function (PMF) of Y?
The Core Idea
Imagine you have a discrete random variable X with a known PMF. Now you transform it using some function g, creating Y = g(X). To find the PMF of Y:
- Identify all possible values of Y: These come from applying g to every value in the range of X
- Group the X-values: For each y-value, find which x-values map to it (i.e., solve g(x) = y)
- Add the probabilities: The probability that Y equals y is the sum of all probabilities P(X = x) where g(x) = y
Mathematically:
P(Y = y) = P(g(X) = y) = Σ P(X = x) for all x where g(x) = y
Example Analogy
Think of X as items with price tags. The function g might be "apply a 20% discount." Some different original prices might result in the same discounted price. To find the probability of a particular discounted price, you add up the probabilities of all original prices that produce that discount.
When It's Simple
If g is one-to-one (each x maps to a unique y), then for each y there's exactly one x, and P(Y = y) = P(X = x) where x = g⁻¹(y).
If g is many-to-one (multiple x-values map to the same y), you must sum all the corresponding probabilities.
Key Takeaway: To find the PMF of Y = g(X) for discrete X, identify each possible y-value and sum the probabilities of all x-values that map to it through g.