Introduction to Discrete Random Variables
Welcome to one of the most practical chapters in your Further Maths journey! In this section of the Statistics Major, we explore how to model events where the outcomes are "countable" (discrete). Whether you are counting the number of goals scored in a match, the number of attempts it takes to pass a level in a game, or how many emails you get in an hour, Discrete Random Variables are the tools you need.
Don't worry if these concepts feel a bit abstract at first. We will break them down into simple steps, using everyday logic and clear formulas. By the end of this, you’ll be able to predict averages and spreads for all sorts of real-world scenarios!
1. Probability Distributions: The Basics
A Random Variable (usually written as \( X \)) is a quantity whose value depends on the outcome of a random event. If it is discrete, it can only take specific, separate values (like 1, 2, 3... but not 1.57).
A Probability Distribution is simply a list, table, or graph showing all possible values of \( X \) and the probability of each one occurring.
Key Rules to Remember:
1. Every individual probability must be between 0 and 1: \( 0 \le P(X=x) \le 1 \).
2. The Golden Rule: All probabilities in a distribution must add up to exactly 1.
\( \sum P(X=x) = 1 \)
Quick Review: If you are given a table with one missing probability, just add up the others and subtract from 1. It’s that simple!
2. Expectation (The Mean) and Variance
When we look at a distribution, we often want to know two things: What is the "average" result, and how "spread out" are the results?
Expectation \( E(X) \)
The Expectation is the long-term average value of the random variable. It is also called the Mean (\( \mu \)).
How to calculate it: Multiply each value by its probability and add them all up.
\( E(X) = \sum x P(X=x) \)
Variance \( Var(X) \)
The Variance measures the spread of the data. It tells us how far, on average, the values are from the mean.
The Formula: \( Var(X) = E(X^2) - [E(X)]^2 \)
To find \( E(X^2) \), you square each \( x \) value, multiply by its probability, and sum them up.
Standard Deviation: This is just the square root of the variance: \( \sigma = \sqrt{Var(X)} \).
Common Mistake to Avoid: When calculating variance, many students forget to square the mean before subtracting it. Remember the "Mean Square minus Square Mean" rule!
Coding (Linear Transformations)
Sometimes we change our variable (e.g., doubling it or adding a constant). Here is how the mean and variance react:
1. \( E(aX + b) = aE(X) + b \) (The mean changes exactly how you’d expect).
2. \( Var(aX + b) = a^2Var(X) \) (Adding \( b \) doesn't change the spread, but multiplying by \( a \) increases the variance by \( a^2 \)).
Key Takeaway:
Expectation is the "center" of your data, and Variance is the "wiggle room."
3. The Discrete Uniform Distribution
This is the simplest model. We use it when every outcome has the exact same probability.
Example: Rolling a fair six-sided die. Every number from 1 to 6 has a probability of \( \frac{1}{6} \).
For a uniform distribution over the values \( \{1, 2, ..., n\} \):
- Mean: \( E(X) = \frac{n+1}{2} \)
- Variance: \( Var(X) = \frac{n^2-1}{12} \)
Did you know? You can use these formulas for any set of consecutive integers (like 4, 5, 6, 7, 8, 9) by adjusting them, but the syllabus specifically expects you to know them for the \( 1 \) to \( n \) case.
4. The Binomial Distribution \( X \sim B(n, p) \)
You’ve seen this in A Level Maths, but it’s vital here too. Use this when you have a fixed number of trials (\( n \)) and only two outcomes: success or failure.
Properties:
- Mean: \( E(X) = np \)
- Variance: \( Var(X) = np(1-p) \)
Memory Aid: Think of "np" as the number of trials times the probability of success. If you take a 10-question multiple-choice test by guessing (\( p=0.25 \)), you’d "expect" to get \( 10 \times 0.25 = 2.5 \) questions right.
5. The Poisson Distribution \( X \sim Po(\lambda) \)
The Poisson Distribution models the number of events happening in a fixed interval of time or space.
When to use it:
You need three conditions for a Poisson model to be valid:
1. Events occur independently.
2. Events occur at a constant average rate (\( \lambda \)).
3. Events occur randomly and singly.
The "Poisson Magic" Property:
In a Poisson distribution, the Mean is equal to the Variance!
\( E(X) = \lambda \)
\( Var(X) = \lambda \)
Quick Tip: If a question asks if a Poisson model is suitable for a dataset, check if the mean and variance are similar. If they are very different, Poisson is likely a bad choice!
Adding Poissons:
If \( X \sim Po(\lambda) \) and \( Y \sim Po(\mu) \) are independent, then their sum is also Poisson:
\( X + Y \sim Po(\lambda + \mu) \)
6. The Geometric Distribution \( X \sim Geo(p) \)
The Geometric Distribution models the "waiting time." It counts the number of trials up to and including the first success.
Example: How many times do I have to flip a coin until I get my first 'Heads'?
Key Formulas:
- Probability of success on the \( r \)-th trial: \( P(X=r) = (1-p)^{r-1}p \)
- Probability of needing more than \( r \) trials: \( P(X > r) = (1-p)^r \)
- Mean (Expectation): \( E(X) = \frac{1}{p} \)
- Variance: \( Var(X) = \frac{1-p}{p^2} \)
Common Mistake: Be careful with "more than" vs "at least." If \( X \sim Geo(0.2) \), the average number of tries to succeed is \( 1 / 0.2 = 5 \).
7. Linear Combinations of Independent Variables
This is where Further Maths steps up a gear. What happens if we combine two different independent random variables, \( X \) and \( Y \)?
Expectation Rules:
Expectations are very friendly. They just add or subtract normally:
\( E(X + Y) = E(X) + E(Y) \)
\( E(X - Y) = E(X) - E(Y) \)
Variance Rules (The Tricky Part!):
Variances always add if the variables are independent, even if you are subtracting the variables themselves!
\( Var(X + Y) = Var(X) + Var(Y) \)
\( Var(X - Y) = Var(X) + Var(Y) \)
Analogy: Imagine two people measuring a room. Person A makes a small error (\( X \)) and Person B makes a small error (\( Y \)). Whether they add their measurements together or subtract them, the total "uncertainty" (variance) in their final answer increases. Errors never cancel each other out!
Key Takeaway:
For independent variables, \( Var(aX \pm bY) = a^2Var(X) + b^2Var(Y) \). Always square the coefficients and always add the variances.
Summary Checklist
Before you move on, make sure you can:
- Use \( \sum P = 1 \) to find missing probabilities.
- Calculate \( E(X) \) and \( Var(X) \) from a table.
- Apply \( E(aX+b) \) and \( Var(aX+b) \).
- Recognize and use the Uniform, Binomial, Poisson, and Geometric models.
- Combine independent variables using the addition rule for variances.
Don't worry if this seems tricky at first! Statistics is all about practice. Once you start recognizing the "story" behind the numbers (e.g., "Oh, this is a waiting time problem, it must be Geometric!"), the formulas will fall into place.