Normal Distribution Properties

Why This Matters

The normal distribution isn't just another statistical concept—it's the foundation for nearly every business decision that involves data. You're being tested on your ability to recognize why data behaves predictably, how to standardize and compare different datasets, and when to apply probability calculations in real-world scenarios. From quality control to financial modeling, the normal distribution shows up everywhere because so many natural and business processes cluster around an average with predictable variation.

Don't just memorize that 68% of data falls within one standard deviation—understand what that means for business decisions. Can you explain why the Central Limit Theorem lets us make inferences from samples? Do you know how to convert any normal distribution to a z-score for comparison? These conceptual connections are what separate students who ace the exam from those who struggle with application questions. Master the underlying logic, and the formulas will make sense.

---

Foundational Shape and Symmetry

The normal distribution's distinctive bell shape isn't arbitrary—it emerges from the mathematical behavior of random variation around a central tendency. These properties define what makes a distribution "normal" in the first place.

Bell-Shaped Symmetry

• The curve is perfectly symmetric around the mean—the left and right halves are exact mirror images of each other
• Symmetry means equal probability on both sides—the chance of falling above the mean equals the chance of falling below
• Skewness equals zero in a true normal distribution, which becomes a key diagnostic when checking if your data is normally distributed

Mean, Median, and Mode Equality

• All three measures of central tendency converge at the center—this only happens in perfectly symmetric distributions
• Simplifies business interpretation because you don't need to debate which "average" to report
• A quick normality check—if mean, median, and mode diverge significantly, your data likely isn't normally distributed

---

Parameters That Define the Distribution

Every normal distribution is completely specified by just two numbers. Understanding how these parameters control the curve's behavior is essential for modeling business data.

Defined by Mean and Standard Deviation

• The mean ($\mu$) sets the center position—shifting the mean slides the entire curve left or right along the x-axis
• The standard deviation ($\sigma$) controls the spread—larger $\sigma$ creates a wider, flatter curve; smaller $\sigma$ creates a taller, narrower curve
• Same mean, different spreads produce distributions that look completely different, which matters when comparing variability across business units

Standard Normal Distribution (Z-Score)

• The standard normal has $\mu = 0$ and $\sigma = 1$—it's the "universal translator" for all normal distributions
• Z-scores measure distance from the mean in standard deviation units—calculated as $z = \frac{x - \mu}{\sigma}$
• Enables cross-comparison between distributions with different scales—comparing sales performance across regions with different averages, for example

---

The Empirical Rule and Probability Structure

These properties give you the practical tools to quantify probability and make predictions. The empirical rule is your mental shortcut; the formal functions give you precision.

68-95-99.7 Rule (Empirical Rule)

• 68% of data falls within $\pm 1\sigma$ of the mean—roughly two-thirds of your observations cluster near the center
• 95% falls within $\pm 2\sigma$—this is why "two standard deviations" often defines "unusual" in business contexts
• 99.7% falls within $\pm 3\sigma$—observations beyond this range are rare enough to flag as potential outliers or errors

Area Under the Curve Equals 1

• Total probability sums to exactly 1 (100%)—this is the fundamental axiom that makes probability calculations work
• Partial areas represent probabilities—the area between any two x-values equals the probability of falling in that range
• The curve never touches the x-axis—it extends infinitely in both directions, though probabilities become negligibly small

Probability Density Function (PDF)

• The PDF formula is $f(x) = \frac{1}{\sigma\sqrt{2\pi}} e^{-\frac{1}{2}\left(\frac{x-\mu}{\sigma}\right)^2}$—you likely won't need to memorize this, but recognize it
• Height at any point shows relative likelihood—taller regions are more probable than shorter regions
• Area under the curve over an interval gives actual probability—the PDF itself doesn't give probability directly for continuous variables

Cumulative Distribution Function (CDF)

• The CDF gives $P(X \leq x)$—the probability of being at or below a specific value
• Always increases from 0 to 1—it's a non-decreasing function that approaches 1 as x approaches infinity
• Most practical for business questions—"What's the probability sales exceed $50,000?" requires CDF calculations (often via z-tables)

---

Powerful Theoretical Properties

These properties explain why the normal distribution dominates statistical inference. They're the theoretical backbone for hypothesis testing, confidence intervals, and business forecasting.

Linear Combinations of Normal Distributions

• The sum of independent normal variables is also normal—if $X \sim N(\mu_1, \sigma_1^2)$ and $Y \sim N(\mu_2, \sigma_2^2)$, then $X + Y \sim N(\mu_1 + \mu_2, \sigma_1^2 + \sigma_2^2)$
• Enables portfolio risk modeling—you can combine individual asset distributions to model total portfolio behavior
• Simplifies complex business systems—multi-step processes with normal variation at each step still yield normal outcomes

Central Limit Theorem

• Sample means approach normality as $n$ increases—regardless of the original population's shape
• Works surprisingly fast—even with $n \geq 30$, the sampling distribution is approximately normal for most populations
• Foundation of statistical inference—this is why we can construct confidence intervals and perform hypothesis tests using normal distribution methods

---

Quick Reference Table

---

Self-Check Questions

1. Which two properties together explain why the normal distribution is symmetric? (Hint: Think about shape and central tendency.)

2. If you have two normal distributions with the same mean but different standard deviations, how would their curves compare visually? What business scenario might require this comparison?

3. Compare and contrast the PDF and CDF: When would you use each one to answer a probability question about sales data?

4. A company's daily revenue follows a normal distribution with $\mu = \$10,000$ and $\sigma = \$1,500$. Using the empirical rule, what range captures approximately 95% of daily revenues? What z-score corresponds to a $13,000 day?

5. FRQ-style: Explain why the Central Limit Theorem allows a business analyst to use normal distribution methods when analyzing average customer spending, even if individual spending amounts are right-skewed. What minimum sample size is typically recommended?