Welcome to Discrete Random Variables!
In H2 Mathematics, you mastered the Binomial distribution. Now, in Further Mathematics, we are going to expand your toolkit with two specialized "counting" models: the Poisson Distribution and the Geometric Distribution. These are essential for modeling everything from how many emails you receive in an hour to how many tries it takes to finally win a game. Don't worry if these look intimidating; we'll break them down step-by-step!
1. The Poisson Distribution \(Po(\mu)\)
Imagine you are standing on a street corner counting how many red cars pass by. You don't know the "maximum" number of cars that could pass (unlike the Binomial distribution where you have a fixed number of trials). You only know the average rate. This is where the Poisson distribution shines.
When is it a suitable model?
For a situation to follow a Poisson distribution, four conditions must be met. You can remember them with the acronym CRIS:
C – Constant rate: The average number of occurrences per unit interval (time or space) stays the same.
R – Random/Independent: One event happening doesn't make the next one more or less likely.
I – Individually: Events cannot occur simultaneously (at the exact same micro-second).
S – Singly: Events occur one at a time.
The Formula
If \(X \sim Po(\mu)\), where \(\mu\) is the mean rate of occurrence:
\(P(X = x) = \frac{e^{-\mu} \mu^x}{x!}\) for \(x = 0, 1, 2, ...\)
Note: The variable \(x\) can go on forever to infinity!
Mean and Variance
The Poisson distribution has a very quirky and helpful property:
Mean: \(E(X) = \mu\)
Variance: \(Var(X) = \mu\)
Quick Review: If you calculate the mean and variance of some data and they are almost equal, it's a huge hint that a Poisson model might be a good fit!
The Additive Property
If you have two independent Poisson variables, you can simply add their rates together!
If \(X \sim Po(\mu_1)\) and \(Y \sim Po(\mu_2)\), then:
\(X + Y \sim Po(\mu_1 + \mu_2)\)
Example: If you get 2 emails per hour on average and your friend gets 3, together you get 5 per hour.
Key Takeaway: The Poisson distribution models the number of occurrences in a fixed interval of time or space.
2. The Geometric Distribution \(Geo(p)\)
While Poisson counts "how many," the Geometric distribution asks "how long until the first success?" Think about trying to roll a "6" on a die. You keep rolling and rolling until you finally succeed. The number of trials you need is your Geometric random variable.
When is it a suitable model?
The conditions for the Geometric distribution are very similar to the Binomial distribution:
1. Trials are independent.
2. There are only two outcomes: Success or Failure.
3. The probability of success, \(p\), is constant for every trial.
4. We stop exactly when the first success occurs.
The Formula
If \(X \sim Geo(p)\), where \(p\) is the probability of success:
\(P(X = x) = (1-p)^{x-1} p\) for \(x = 1, 2, 3, ...\)
Analogy: To succeed on the 10th try, you must fail 9 times first, then succeed on the 10th. So, \(P(X=10) = (1-p)^9 \times p\).
Mean and Variance
These formulas are slightly more complex, but very important:
Mean (Expected value): \(E(X) = \frac{1}{p}\)
Variance: \(Var(X) = \frac{1-p}{p^2}\)
Did you know? If the chance of winning a game is \(1/10\), the mean tells us you should expect to play 10 times to get your first win. It's just the reciprocal!
Key Takeaway: The Geometric distribution models the number of trials until the first success happens.
3. Common Pitfalls to Avoid
1. Wrong Interval for Poisson: If the rate is 3 per hour, but the question asks for the probability in 2 hours, you must change \(\mu\) to 6. Always match your \(\mu\) to the time/space interval in the question!
2. Starting at Zero: In the Poisson distribution, \(X\) can be 0 (nothing happened). In the Geometric distribution, \(X\) must be at least 1 (you have to try at least once to succeed).
3. Independence: You cannot add Poisson variables or use the Geometric distribution if the events affect each other. Always check if the question implies independence.
Summary Checklist
Before you move on to practice questions, make sure you can:
- [ ] State the conditions for Poisson (CRIS) and Geometric models.
- [ ] Calculate probabilities for \(P(X=x)\) for both distributions.
- [ ] Recall that for Poisson, Mean = Variance.
- [ ] Use the Additive Property for Poisson to combine independent rates.
- [ ] Calculate the expected number of trials (\(1/p\)) for a Geometric distribution.
Pro-tip: Don't worry if the formulas seem tricky at first. Most students find that after practicing 5 or 10 questions, the "rhythm" of these distributions becomes second nature!