Scribd
Upload
10 views12 pages

Lecture 5

The lecture discusses the normal random variable and its distribution, highlighting key concepts such as the density curve, cumulative distribution function, and the standard normal distribution. It introduces the normal distribution formula, the 68-95-99.7 rule, and provides examples to illustrate standardization and probability calculations. The importance of the normal distribution in statistics is emphasized, along with its unique properties and applications.

Uploaded by

Raymond Wu
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
10 views12 pages

Lecture 5

The lecture discusses the normal random variable and its distribution, highlighting key concepts such as the density curve, cumulative distribution function, and the standard normal distribution. It introduces the normal distribution formula, the 68-95-99.7 rule, and provides examples to illustrate standardization and probability calculations. The importance of the normal distribution in statistics is emphasized, along with its unique properties and applications.

Uploaded by

Raymond Wu
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 12

Lecture 5: Normal Random Variable and Distribution

STOR 435, Spring, 2025

1/21/2025

435-Spring-2025 normal
Bell curve 1

Histogram vs Density Curve

435-Spring-2025 normal
density

Continuous RV (random variable) X


Density curve f (x) for X that satisfies:

f (x) ≥ 0, ∀ x ∈ IR
Z ∞
f (x) dx = 1
−∞
Z b
P(a ≤ X ≤ b) = f (x) dx
a
(interval probability)

435-Spring-2025 normal
density continued

Infinitesimal probability P(X ∈ dx) ≈ f (x) dx, note the abuse


of notation here: dx represents a tiny neighborhood of x
and also the length of such a neighborhood.
cdf (cumulative distribution function)
Z x
F(x) = f (u) du, x ∈ IR
−∞

dF(x) dF(x)
dx = f (x), ∀ x at which dx exists.
Note: For a continuous RV X, P(X = x) = 0, ∀ x
(no atoms!)

435-Spring-2025 normal
normal distribution

X ∼ N(µ, σ 2 ): µ = mean (location), σ 2 = variance (scale),


or σ = standard deviation. The density:
1 − 1
(x−µ)2
f (x) = √ e 2σ 2 , x ∈ IR
2πσ 2
where µ ∈ IR and σ > 0 are two important (constant)
parameters.

435-Spring-2025 normal
Bell curve 2

Example: The normal distribution is the most important


distribution in Statistics. Typical normal curves with
different sigma (standard deviation) values are shown
below.

435-Spring-2025 normal
standard normal

Standard normal distribution: Z ∼ N(0, 1) with density


1 2
ϕ(z) = √12π e− 2 z and cdf
Z z
Φ(z) = ϕ(u) du
−∞
.
Denote Φ(a, b) = Φ(b) − Φ(a) and note that
Φ(−1, 1) ≈ 0.6826, Φ(−2, 2) ≈ 0.9544 Φ(−3, 3) ≈ 0.9975.

435-Spring-2025 normal
Bell curve 3

The 68-95-99.7 Rule


2 3

435-Spring-2025 normal
standardization

X−µ
Fact: X ∼ N(µ, σ 2 ) ⇔ Z = σ ∼ N(0, 1).
If X ∼ N(µ, σ 2 ), then
   
b−µ a−µ
P(a ≤ X ≤ b) = Φ −Φ
σ σ

Although each pair (µ, σ) defines a distinct member in the


normal family, there is only one normal table based on
N(0, 1).

435-Spring-2025 normal
Example 1

Assume X ∼ N(µ, σ12 ), Y ∼ N(µ, σ22 ), and


P(|X − µ| > 2) > P(|Y − µ| > 2).
Is σ1 < σ2 ? Why?

Answer: False. Use standardization.

435-Spring-2025 normal
Example 2

Suppose X ∼ N(µ, σ 2 ), P(X < 0) = 0.5 and


P(X ≥ 1) = 0.05. Find P(|X| ≤ 0.25).

Solution: P(X < 0) = 0.5 ⇒ µ = 0.


P(X ≥ 1) = P(Z ≥ 1/σ) = 0.05 ⇒ 1/σ = 1.645. Hence

P(|X| ≤ 0.25) = P(|Z| ≤ 0.25/σ) = · · · = 0.3182.

435-Spring-2025 normal
Example 3

Define a function g(t) = P(t ≤ Z ≤ t + 2), t ∈ IR, where


Z ∼ N(0, 1). Find the value t that maximizes g(t).

Note that g(t) = Φ(t + 2) − Φ(t). Setting


dg(t)
dt = ϕ(t + 2) − ϕ(t) = 0 will lead to t = −1. (why?) This
can also be interpreted based on the N(0, 1) curve.

435-Spring-2025 normal

You might also like