1. Definition
- The normal distribution (also called Gaussian distribution) is a continuous probability distribution shaped like a bell curve.
- Many natural and social phenomena (heights, test scores, measurement errors) approximately follow it.
2. Probability Density Function (PDF)
For a random variable $X \sim N(\mu, \sigma^2)$:
$f(x) = \frac{1}{\sigma \sqrt{2\pi}} \exp\!\Bigg(-\frac{(x-\mu)^2}{2\sigma^2}\Bigg)$
Where:
- $\mu$ = mean (center of the distribution).
- $\sigma^2$ = variance (spread).
- $\sigma$ = standard deviation.
3. Properties
- Symmetry: Bell-shaped and symmetric around the mean $\mu$.
- Mean, Median, Mode: all equal to $\mu$.
- Spread: Controlled by $\sigma$. Larger $\sigma$ = flatter and wider curve.
- Total probability = 1 (area under curve).
4. Standard Normal Distribution
- Special case: $\mu = 0, \sigma = 1$.
- Denoted $Z \sim N(0,1)$.
- Used with z-scores: $z = \frac{x – \mu}{\sigma}$ to standardize data and use standard normal tables.
5. Probability Ranges (Empirical Rule: 68–95–99.7 Rule)
- About 68% of values lie within ±1σ of the mean.
- About 95% within ±2σ.
- About 99.7% within ±3σ.
6. Examples
- Human height: ~ Normal(170, 10²).
- Test scores (after standardization).
- Measurement errors in experiments.
7. Applications
- Statistics: hypothesis testing, confidence intervals, regression assumptions.
- Forecasting: modeling demand, returns, or prediction errors.
- Machine Learning: Gaussian likelihoods, loss functions, Bayesian priors.
8. Visualization (Conceptual)
- Small σ → tall, narrow curve.
- Large σ → flat, wide curve.
- Symmetric about $\mu$.
Summary:
The Normal Distribution $N(\mu, \sigma^2)$ is a bell-shaped, symmetric distribution described by mean $\mu$ and variance $\sigma^2$. It is central in probability and statistics due to the Central Limit Theorem, which says sums/averages of many random variables tend to be approximately normal.
