Welcome to the World of Continuous Random Variables!

In your previous studies, you likely looked at Discrete Random Variables—things you can count, like the number of heads in a coin toss or the number of students in a class. But what happens when we measure things like height, time, or weight? These things don't "jump" from one number to the next; they flow along a scale. This is the world of Continuous Random Variables.

In this chapter, we focus on the most important continuous distribution in the world: the Normal Distribution. Mastering this is like finding a "skeleton key" that unlocks many doors in statistics!


1. Discrete vs. Continuous: What’s the Difference?

Before we dive into the math, let's make sure we understand what makes a variable "continuous."

  • Discrete Random Variables: These have specific, separate values. Example: Counting the number of goals in a soccer match (you can't score 2.45 goals).
  • Continuous Random Variables: These can take any value within a certain range. They are usually measurements. Example: The time it takes for you to run 100 meters (it could be 12 seconds, 12.1 seconds, or even 12.1045 seconds).

Did you know? Because a continuous variable has infinite possible values, the probability of the variable being exactly one specific number (like exactly 1.750000... meters tall) is actually zero! Instead, we always look for the probability that a value falls within a range.


2. The Normal Distribution: The "Bell Curve"

The Normal Distribution is a special type of continuous random variable. It is used to model things that cluster around an average, like exam scores or the lengths of leaves on a tree.

Key Features:

  • Symmetrical: The left side is a mirror image of the right side.
  • Bell-Shaped: Most of the data is in the middle, and it tapers off at the ends.
  • The Mean (\(\mu\)): This is the center of the curve.
  • The Variance (\(\sigma^2\)): This tells us how "spread out" the bell is.

We write it as: \(X \sim N(\mu, \sigma^2)\)

Example: If the average height of students is 160cm with a variance of 25, we write \(X \sim N(160, 25)\). Note: Here, the standard deviation (\(\sigma\)) is \(\sqrt{25} = 5\).

Quick Review: Always check if the question gives you the variance (\(\sigma^2\)) or the standard deviation (\(\sigma\)). In the formula \(X \sim N(\mu, \sigma^2)\), the second number is the variance!


3. Standardizing: Using Z-Scores

Every Normal curve is different (some are tall and thin, others short and wide). To find probabilities, we need a way to compare them all to one "standard" version called the Standard Normal Distribution, which has a mean of 0 and a variance of 1.

We do this using the Z-score formula:

\(Z = \frac{X - \mu}{\sigma}\)

Analogy: Think of the Z-score as a Universal Translator. Just as you might convert different currencies (Dollars, Euros, Pesos) into Gold to compare their value, we convert different Normal variables into Z-scores to compare them.

Steps to Standardize:

  1. Take your value (\(X\)).
  2. Subtract the mean (\(\mu\)).
  3. Divide by the standard deviation (\(\sigma\)).

Don't worry if this seems tricky at first! Just remember: The Z-score tells you how many "standard deviations" your value is away from the average.


4. Finding Probabilities Using Tables

Once you have a Z-score, you use the Normal Distribution Table to find the probability. The table tells you the area to the left of a Z-score, which we call \(\Phi(z)\).

Common Scenarios:

  • To find \(P(Z < a)\): Just look up \(a\) in the table.
  • To find \(P(Z > a)\): Use symmetry! The total area is 1, so \(P(Z > a) = 1 - \Phi(a)\).
  • To find \(P(a < Z < b)\): Find the area to the left of \(b\) and subtract the area to the left of \(a\). Formula: \(\Phi(b) - \Phi(a)\).

Memory Aid: If you want the area to the Right, you must Subtract from 1. (Right = Remove from 1).


5. Inverse Problems: Finding \(\mu\) or \(\sigma\)

Sometimes the examiner gives you the probability (the area) and asks you to find the value of \(X\), the mean, or the standard deviation. This is like "working backward."

The Step-by-Step Process:

  1. Draw a sketch of the bell curve and shade the given area.
  2. Use the table "inside-out" to find the Z-score that matches the probability.
  3. Plug the Z, \(\mu\), and \(\sigma\) into the formula \(Z = \frac{X - \mu}{\sigma}\).
  4. Solve for the missing letter.

Common Mistake: If the area is on the left side of the mean (less than 50%), your Z-score must be negative. The table usually only gives positive Z-scores, so you'll need to use symmetry!


6. The Normal Approximation to the Binomial

Sometimes we have a Binomial Distribution (like flipping a coin 1,000 times), but calculating it exactly is too hard. If the sample size is large enough, the Binomial distribution starts to look like a Normal curve!

When can you use this?

You can use the Normal approximation only if:

  • \(np > 5\)
  • \(n(1-p) > 5\) (also written as \(nq > 5\))

How to do it:

  1. Calculate the Mean: \(\mu = np\).
  2. Calculate the Variance: \(\sigma^2 = np(1-p)\).
  3. Use Continuity Correction (explained below).
  4. Standardize and find the probability as usual.

7. The Continuity Correction: The "0.5 Rule"

When we move from a Discrete variable (bars) to a Continuous variable (a smooth line), we have to account for the space between the numbers. This is called the Continuity Correction.

Think of each discrete number as a "block" that spans 0.5 units on either side. For example, the number 10 actually covers the space from 9.5 to 10.5.

How to Adjust:

  • If you want \(P(X \le 10)\), you include the whole block for 10, so you use 10.5.
  • If you want \(P(X < 10)\), you don't include 10, so you stop at 9.5.
  • If you want \(P(X \ge 10)\), you start at the beginning of the block, which is 9.5.
  • If you want \(P(X > 10)\), you start after the block, which is 10.5.

Key Takeaway: Always draw a quick number line! If you want to include the number, stretch your boundary out by 0.5. If you want to exclude it, pull your boundary back by 0.5.


Summary Checklist

Before you sit your exam, make sure you can:

  • Identify the difference between discrete and continuous variables.
  • Standardize any value using \(Z = \frac{X - \mu}{\sigma}\).
  • Read the Normal Distribution table correctly (including for negative Z-values).
  • Work backward from a probability to find a value.
  • Check conditions (\(np > 5\) and \(nq > 5\)) for the Normal approximation.
  • Apply the 0.5 continuity correction accurately.

You've got this! Statistics is just a series of logical steps. Take it one step at a time, draw your diagrams, and the answers will follow.