1. Continuous Uniform Distribution

Definition — Uniform Distribution

has constant density on :

Theorem — Mean and Variance

Proof. .
, so .

Worked Example

A bus arrives at a uniformly random time in a 30-minute window, .

  • min, , min

⚠️ For a uniform RV, probability = (length of interval inside ) / . Clip the interval to first.


2. Normal (Gaussian) Distribution

Definition — Normal Distribution

has density

Properties of the Normal Curve

  1. Bell-shaped, symmetric about ; mean = median = mode .
  2. Inflection points at .
  3. Approaches the axis asymptotically in both directions.
  4. Total area under the curve is 1.

Theorem — The Density Integrates to 1

Proof. Substitute ; it suffices to show . In polar coordinates,

Theorem — Mean and Variance


(The mean follows from symmetry; the variance from integration by parts on .)


3. Areas Under the Normal Curve — Standardization

Theorem — Standardizing

If then

and

where is read from the standard normal table.

Proof. ; substitute , , to get .

Useful Identities

Empirical (68–95–99.7) Rule

IntervalProbability
0.6827
0.9545
0.9973

Worked Example — Forward Lookup

Line voltage . Find .

N(μ=50, σ=10); shaded P(45 < X < 62) = 0.5764 0.5764 20 30 40 45 50 62 70 80

Worked Example — Reverse Lookup (percentile)

Find with . From the table, (more precisely ):

⚠️ Two different questions: given , find a probability (, then the table) vs. given a probability, find (look up inside the table body, then ).

⚠️ Notation varies: some texts write with the variance. Check whether “N(50, 100)” means or .


4. Applications of the Normal Distribution

Example — Tolerance. Resistors have ; spec limits are .

About 13.4% of resistors are out of spec.

Example — Grading cut-off. Exam scores . The top 10% get an A: cut-off .


5. Normal Approximation to the Binomial

Theorem — De Moivre–Laplace

If then, as ,

(A special case of the Central Limit Theorem, Chapter 10, since is a sum of Bernoulli variables.)

Rule of thumb: good when and .

Definition — Continuity Correction

A discrete value corresponds to the interval under the continuous curve:

BinomialNormal approximation
=

Worked Example

, : , . Find .

MethodResult
Exact binomial0.1795
Normal with continuity correction0.1792
Normal without correction ()0.1537
Binomial(100, 0.4) bars with N(40, 4.899) curve; shaded bars = X ≤ 35 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54

⚠️ Forgetting the continuity correction is the most common error here; it matters most when is moderate.