Welcome to Discrete Probability Distributions!
In your standard A Level Maths course, you’ve already met the basics of probability. Now, in Further Statistics 1, we are going to take those ideas and supercharge them. We aren’t just looking at the chance of something happening; we are looking at the "long-term behavior" of systems. Whether you are predicting the number of defective items in a factory or the average points scored in a game, these tools are your best friend.
Don't worry if this seems tricky at first! We are going to break it down step-by-step, starting with the most important concepts: the Mean and the Variance.
1. The Expected Value: \(E(X)\)
The Expected Value (often written as \(E(X)\) or the Greek letter \(\mu\)) is just a fancy way of saying the Mean. If you ran an experiment thousands of times, the Expected Value is the average result you would get.
The Formula
\(E(X) = \mu = \sum xP(X=x)\)
How to calculate it (Step-by-Step)
1. Take each possible value of \(x\).
2. Multiply that value by its probability, \(P(X=x)\).
3. Add all those results together!
Example: Imagine a game where you have a 0.2 chance of winning £10 and a 0.8 chance of winning £0.
\(E(X) = (10 \times 0.2) + (0 \times 0.8) = £2\).
This doesn't mean you will ever actually win £2 in a single game (you either win £10 or £0!). It means that if you played many times, you would average a win of £2 per game.
Quick Review: The Expected Value is the average outcome in the long run.
2. Variance: \(Var(X)\)
While the Mean tells us the "center," the Variance (written as \(Var(X)\) or \(\sigma^2\)) tells us how much the data "wiggles" or spreads out from that center. A high variance means the results are all over the place; a low variance means they are clustered close to the mean.
The Formula
\(Var(X) = \sigma^2 = E(X^2) - [E(X)]^2\)
In a more expanded form, this looks like:
\(Var(X) = \sum x^2P(X=x) - \mu^2\)
Step-by-Step Breakdown
1. Find \(E(X)\) first: This is your mean (\(\mu\)).
2. Calculate \(E(X^2)\): Square each \(x\) value, multiply by its probability, and sum them up.
3. Subtract the square of the mean: Take your result from step 2 and subtract \(\mu^2\).
Common Mistake to Avoid: Many students forget to square the mean at the end! Remember: "Mean of the squares MINUS square of the mean."
Key Takeaway: Variance measures consistency. If you are a baker, you want the variance of the weight of your loaves to be very low so every customer gets the same size bread!
3. Functions of a Random Variable: \(E(g(X))\)
Sometimes, we don't just want the mean of \(X\). We might want the mean of \(X^2\), or \(3X + 2\), or any other function. We call this \(E(g(X))\), where \(g(x)\) is just a rule we apply to our data.
The Rule
To find the expected value of a function, you apply the function to the \(x\) values, but keep the probabilities the same.
\(E(g(X)) = \sum g(x)P(X=x)\)
Analogy: Imagine a taxi company. \(X\) is the number of miles a passenger travels. The fare is calculated as \(g(X) = 2X + 5\). To find the expected fare, you don't need a new probability table; you just apply the fare rule to the mileage values.
Special Shortcuts (Expectation Algebra)
If \(a\) and \(b\) are constants:
1. \(E(aX + b) = aE(X) + b\)
2. \(Var(aX + b) = a^2Var(X)\)
Did you know? Adding a constant (like \(b\)) doesn't change the variance! If everyone in a class gets a 5-mark bonus, the average goes up, but the spread of the marks stays exactly the same.
4. Assessing the Suitability of a Model
In Further Maths, you won't just do calculations; you’ll be a "Maths Detective." You might be given a real-world situation and asked if a specific discrete distribution (like a Discrete Uniform Distribution) is a good fit.
How to decide if a model is suitable:
1. Calculate the theoretical mean and variance based on the model (e.g., if every outcome is supposed to be equally likely).
2. Compare these to the experimental mean and variance from the actual data.
3. Conclusion: If the numbers are close, the model is a good fit. If the variance of the data is much higher than the model predicts, the model is likely unsuitable.
Key Takeaway: Models are just simplified versions of reality. If the "maths" of the model doesn't match the "facts" of the data, we need a better model!
Summary: The Essentials
Prerequisite Check: Make sure you are comfortable with the fact that all probabilities in a distribution must sum to 1.
Quick Formula Reference:
Mean: \(E(X) = \sum xP(x)\)
Variance: \(Var(X) = E(X^2) - (E(X))^2\)
Function Expectation: \(E(g(X)) = \sum g(x)P(x)\)
Memory Aid: For variance, think "MS - SM" (Mean of Squares minus Square of Mean). It sounds like a secret code, but it will save you marks in the exam!