Mixed Random Variables
What you'll learn: You'll learn how to work with random variables that have both discrete jumps and continuous components in their probability distributions.
When Discrete Meets Continuous
So far, you've seen random variables that are purely discrete (like dice rolls) or purely continuous (like measurements). But real-world scenarios sometimes produce mixed random variables—random variables that behave discretely in some parts of their range and continuously in others.
Understanding the Structure
A mixed random variable has a probability distribution that assigns:
- Discrete probability masses to certain specific values (like a PMF)
- Continuous probability density across intervals (like a PDF)
Think of it like this: imagine a game where you spin a wheel. If the spinner lands exactly on the "jackpot" marker (probability 0.1), you win a fixed 0 and 100 and continuous spread elsewhere.
Describing Probabilities
For a mixed random variable X:
- Use point probabilities P(X = x) for the discrete components
- Use a probability density function f(x) for the continuous regions
- The total probability across all components must sum/integrate to 1
The cumulative distribution function (CDF) of a mixed random variable shows jump discontinuities where the discrete masses occur, and smooth increases where the continuous portions lie.
Example Structure
If X represents waiting time at a bus stop where buses sometimes arrive instantly (10% chance) or uniformly over 20 minutes (90% chance), then:
- P(X = 0) = 0.10 (discrete mass)
- For 0 < x ≤ 20: continuous density proportional to the remaining 90% probability
Key Takeaway: Mixed random variables combine discrete probability masses at specific points with continuous probability density over intervals, requiring both PMF-like and PDF-like descriptions to fully characterize their behavior.