Scribd
Upload
102 views

STOR 435.002 Lecture 10: Continuous Random Variables - II

1) The document discusses normal random variables and their properties including the normal distribution formula, examples of normal distributions in real life, and how to standardize a normal random variable. 2) It also discusses how the normal distribution can be used to approximate binomial distributions for large sample sizes using the central limit theorem. Examples are provided to demonstrate calculating probabilities for normal and binomial distributions. 3) The document concludes by solving an example problem using the normal approximation to the binomial to calculate the probability that a university admits more than their ideal enrollment size, given their acceptance rate.

Uploaded by

quytruyenkiep
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
102 views

STOR 435.002 Lecture 10: Continuous Random Variables - II

1) The document discusses normal random variables and their properties including the normal distribution formula, examples of normal distributions in real life, and how to standardize a normal random variable. 2) It also discusses how the normal distribution can be used to approximate binomial distributions for large sample sizes using the central limit theorem. Examples are provided to demonstrate calculating probabilities for normal and binomial distributions. 3) The document concludes by solving an example problem using the normal approximation to the binomial to calculate the probability that a university admits more than their ideal enrollment size, given their acceptance rate.

Uploaded by

quytruyenkiep
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/ 17

STOR 435.

002 Lecture 10

Continuous Random Variables - II

Ruoyu Wu

UNC Chapel Hill

1 / 17
Normal random variables

Definition
X is a normal random variable with parameters µ and σ 2 if its density is

1 (x−µ)2
−
f (x) = √ e 2σ2 , −∞ < x < ∞.
2πσ 2

Notation: X = N (µ, σ 2 ).

2 / 17
Normal random variables

Note 1:
Importance: many real life distributions follow a normal density curve. We will
see later in Chapter 8 (Lecture 17) that normal distributions arise in an
important result known as Central Limit Theorem.

Examples

1. Standardized test scores: http:


//nextsteptestprep.com/2012/04/13/average-lsat-score/
2. Distribution of Physiological measurements: e.g. Blood pressure, beta
carotene levels etc.
3. Used in ranking individuals in large companies. Stack ranking.
4. As a major driving component in financial models. Black-Scholes
model.
5. Approximations to a wide array of other distributions. The Central Limit
theorem. See end of these slides for a specific example.

3 / 17
Normal random variables

Note 2:
R∞
√ 1 is to have f (x)dx = 1.
2πσ 2 −∞

4 / 17
Normal random variables

Note 3:
If X = N (µ, σ 2 ) and a, b are two numbers, then the random variable
aX + b = N (aµ + b, a2 σ 2 ).

Terminology: N (0, 1) is called a standard normal random variable. Note that,


if X = N (µ, σ 2 ), then
X −µ
= N (0, 1)
σ
is a standard normal. This procedure is called standardizing, and allows all
computations involving N (µ, σ 2 ) to be reduced to those for N (0, 1).

5 / 17
Normal random variables

Note 4:
If X = N (µ, σ 2 ) then µ = EX, σ 2 = V ar(X).

6 / 17
Derivation of moments continued

7 / 17
Normal random variables

8 / 17
Normal random variables

Note 5:
The cumulative distribution function of N (0, 1) is
Z x
1 y2
Φ(x) = √ e− 2 dy, −∞ < x < ∞,
2π −∞
and cannot be computed explicitly.

For x > 0, Φ(x) are given in Table 5.1 of the textbook. For x < 0, use
Φ(x) = 1 − Φ(−x).
9 / 17
Normal random variables

10 / 17
Continuous random variables

Example 4b: If X is a normal random variable with parameters µ = 3 and σ 2 = 9, find


(a) P (2 < X < 5); (b) P (X > 0); (c) P (|X − 3| > 6).

11 / 17
Normal random variables

Normal approximations to binomial


Let Sn = B(n, p) be the number of successes in n independent Bernoulli
trials. Then, ESn = np, V ar(Sn ) = np(1 − p) and, for large n,1
Sn − np
p ≈ N (0, 1).
np(1 − p)

1
The approximation is quite good for np(1 − p) ≥ 10. Also, compare to Poisson
approximations: p is not small here.
12 / 17
Normal Approximation to the Binomial

Continuity correction
Slightly subtle point: Binomial distribution is a discrete distribution while
Normal is a continuous distribution. When approximating probabilities of the
Binomial, need the continuity correction

13 / 17
Schlitz vs Michelbob: Live taste test

See the following advertisement


https://adland.tv/commercials/
schlitz-schlitz-vs-michelob-live-1981-060-usa

More reading
I http://wwnorton.tumblr.com/post/40864025234/
the-naked-statistics-of-the-1981-schlitz-super-bowl-blin
I http://www.nytimes.com/1981/01/14/business/
advertising-schlitz-to-challenge-michelob-s-taste.html

14 / 17
Normal Approximation to the Binomial

Example
Ruoyu is the head of a major PR firm and plans an advertising campaign for a VP beer
company. Leveraging on the fact that for non-specialty beers, most people cannot tell
the difference between different brands of beer, he picks 100 affirmed drinkers of Bond
beer and then does a live test on television where these individuals are given VP and
Bond beer in unmarked mugs (so that we may assume each person will chose VP and
Bond with 0.5 chance). Let X be the number of individuals who prefer VP beer. Find
(a) P (X > 40)
(b) P (40 ≤ X ≤ 50).

15 / 17
Solution contd

16 / 17
Normal Approximation to the Binomial

Example: The ideal size of an awesome university is 4200. The college, knowing from
past experience that on average only .43 percent of those accepted for admission will
actually attend, uses a policy of approving the applications of 9500 students. Compute
the probability that more than (>) 4200 first-year students attend this college.2

2
I used the data from the Fall 2015 admissions profile of UNC.
17 / 17

You might also like