STATS 2B03 - Statistical Methods for

Science

Instructor:
Eman Alamer, PhD.

Department of Mathematics and Statistics

McMaster University
Introduction

Today, we will look at:

Chapter 5- Sections: 5.1, 5.2, 5.3

Chapter 6- Section: 6.1

Concepts: probability distributions, binomial distribution, Poisson

distribution, standard normal distribution.

STATS 2B03 (Lecture 3) 1 / 30 1

5.1: Probability Distributions

A random variable (r.v.) is a variable that has a single numerical

value, determined by chance, for each outcome of an experiment.

A r.v. is discrete if we can list all of its possible values.

A probability distribution is a description that gives the probability

for each value of the random variable. It is often expressed in the
format of a table, formula, or graph.

STATS 2B03 (Lecture 3) 2 / 30 2

Example

If we flip a coin 3 times, identify

1 A discrete random variable

2 The probability distribution

STATS 2B03 (Lecture 3) 3 / 30 3

Mean, Variance of Discrete Random Variable

If a discrete random variable X can take values x1 , x2 , ..., xn with

probabilities p1 , p2 , ..., pn then
The discrete probability distribution
X
p(x) = 1, 0 ≤ p(x) ≤ 1
all x

The mean µ or expected value E(X)

X
µ = E(X) = xp(x)
all x

The variance σ 2 or V ar(X)

X X
σ 2 = V ar(X) = (x − µ)2 p(x) = x2 p(x) − µ2
all x all x

The standard deviation σ √

σ= σ2
STATS 2B03 (Lecture 3) 4 / 30 4
Example

If 3 couples each have one child, let X be the number of girls.

1 Find the discrete probability distribution.

2 Find the mean of X.

3 Find the variance of X.

4 Find the standard deviation of X.

STATS 2B03 (Lecture 3) 5 / 30 5

Identifying Significant Values

Range Rule of Thumb (for identifying significant values):

Values larger than µ + 2σ are called significantly high.

Values smaller than µ − 2σ are called significantly low.

Significant Results with Probabilities:

X successes among n trials is significantly high if the probability of X

or more successes is 0.05 or less. i.e.,P (X or more successes) ≤ 0.05.

X successes among n trials is significantly low if the probability of X

or fewer successes is 0.05 or less. i.e.,P (X or fewer successes) ≤ 0.05

Note: If the probability of a single outcome is small, it does not make

that single outcome significantly low or high.

STATS 2B03 (Lecture 3) 6 / 30 6

Example

Chapter Problem: Before its clinical trials were discontinued, the

Genetics & IVF Institute conducted a clinical trial of the XSORT
method designed to increase the probability of conceiving a female
and, among the 945 babies born to parents using the XSORT
method, there were 879 females....
If 945 couples each have one child, and P (girl) = 12 , µ = 472.5,
σ 2 = 236.25, and σ = 15.37
1 Is 879 of those couples having girls is significantly high ( Use range rule
of thumb )?
2 Is 879 of those couples having girls is significantly high, let
P (879 or more girls) = 0.00000? (Use significant results with
probabilities).
3 Is 474 girls significantly high, let P (474) = 0.0258? (Use significant
results with probabilities)

STATS 2B03 (Lecture 3) 7 / 30 7

Rare Event Rule for Inferential Statistics

Rare Event Rule for Inferential Statistics: If under a given

assumption, an observed outcome is significantly high or significantly
low we conclude that the assumption is probably not correct.

1
Example: If 945 couples each have one child, and P (girl) = 2 then
879 girls is significantly high. What can we conclude?

STATS 2B03 (Lecture 3) 8 / 30 8

Example

In a group of 12 people, 8 have brown eyes, 4 have green eyes. Three

people are chosen at random and 2 of them have green eyes. Is this
significantly high?

1 Use significant results with probabilities

2 Use range rule of thumb

STATS 2B03 (Lecture 3) 9 / 30 9

Section 5.2: The Binomial Distribution

A Bernoulli trial is an experiment with two outcomes ”success” and

”failure”.

Bernoulli experiment called ”binary” or ”dichotomous” experiment.

p denote the probability of a ”success”, q = 1 − p denote the

probability of a ”failure”.

A Bernoulli process (or binomial experiment) consists of

1 n independent Bernoulli trials.

2 The probability of success p is the same for each trial.

STATS 2B03 (Lecture 3) 10 / 30 1

Combinations (Number of Arrangements)

The notation nx or Cxn or n C x represents the number of ways to

select/choose x items from n where the order in the selection doesn’t

matter. x successes from n trials. The formula for a combination is
 
n n!
= Cxn = n C x =
x x!(n − x)!

where n! = n × (n − 1) × .... × 2 × 1

STATS 2B03 (Lecture 3) 11 / 30 1

Example

In how many ways can 3 S’s and 2 F’s be arranged?

STATS 2B03 (Lecture 3) 12 / 30 1

Binomial Distribution

If X is the number of successes in a Bernoulli process, then X is

called a binomial random variable with parameters n and p.
If X is a binomial random variable with parameters n and p then the
probability of obtaining x successes in these n trials, is given by:
 
n x
P (x) = P (x successes) = p (1 − p)n−x , x = 0, 1, ..., n
x

We denoted this by X ∼ Binomial(n, p) or X ∼ Bin(n, p).

We use the binomial distribution when we are interested in the
probability of having x successes out of n trials.
If X is a binomial random variable, then
The mean µ = E(X) = np
The variance σ 2 = V ar(X) = np(1 − p)

STATS 2B03 (Lecture 3) 13 / 30 1

Probability Keywords and Translation

Exactly 5: P(X = 5).

At most 5: P(X ≤ 5).

At least 5: P(X ≥ 5).

Fewer than 5, below 5: P(X < 5).

More than 5, exceed 5: P(X > 5).

Between 3 and 5, inclusive: P(3 ≤ X ≤ 5).

STATS 2B03 (Lecture 3) 14 / 30 1

Example

A multiple choice test has 20 questions, with 5 choice for each

question. Suppose that a student guesses on every question.

1 Find the probability of getting exactly 6 correct answers.

2 Find the probability of passing the test.

3 Find the probability of failing the test.

4 Find the mean.

5 Find the variance.

6 Find the standard deviation.

STATS 2B03 (Lecture 3) 15 / 30 1

Example

A surgeon performs 10 surgeries. Assume surgeries are independent

and each has a 0.1 probability of later developing a complication. Let
X be the number of surgeries that lead to a complication.

1 Find the probability of exactly 2 surgeries develop complications.

2 Find the probability of at most 2 surgeries develop complications.

3 Find the probability of more than 2 surgeries develop complications.

4 Find the mean.

5 Find the variance.

6 Find the standard deviation.

STATS 2B03 (Lecture 3) 16 / 30 1

Example

A fair die is rolled 26 times and lands on 6 once. Is it significantly

low?

STATS 2B03 (Lecture 3) 17 / 30 1

The Poisson Distribution

Let X be the number of occurrences of an event per unit of ”time”,

where there are a large number of possible occurrences of the event,
but the probability of each occurrence is small.

Let λ be the average number of occurrences of the event per unit of

”time”.

Then X is called a Poisson random variable with parameter λ.

A random variable X is the number of successes in a certain amount

of time or space.

Used to model the number of events that occur in an interval of time.

STATS 2B03 (Lecture 3) 18 / 30 1

Poisson Distribution

If X is a Poisson random variable with parameters λ then the

probability that x will occur is given by

λx e−λ
P (x) = P (x ocurrences) = , x = 0, 1, 2, ...
x!
We denoted this by X ∼ Poisson(λ) or X ∼ Poi(λ).

If X is a Poisson random variable, then

The mean µ = E(X) = λ

The variance σ 2 = V ar(X) = λ

STATS 2B03 (Lecture 3) 19 / 30 1

Example

A hospital emergency room receives an average of 5 new patients per

hour. Find the probability that

1 Exactly 8 people will be admitted in the next hour.

2 At least 3 people will be admitted in the next hour.

STATS 2B03 (Lecture 3) 20 / 30 2

Example

Let’s assume a doctor writes on average 15 prescriptions per day. Let

X be the number of prescriptions the doctor writes on a given day.
Find the:

1 Mean (average) number of prescriptions written in one day.

2 Variance of the number of prescriptions written in one day.

3 Standard deviation of the number of prescriptions written in one day.

STATS 2B03 (Lecture 3) 21 / 30 2

Section 6.1 The Standard Normal Distribution

A random variable is continuous if it can assume all values in a given

interval.

Continuous random variables when the random experiments involve

measuring, ”a continuous set of values”.

A continuous random variable takes a range of values ( finite or

infinite). E.g. [0, 1] , (0, ∞), (−∞, ∞), [a, b].

STATS 2B03 (Lecture 3) 22 / 30 2

Continuous Random Variable

A random variable X is continuous if there is a function f (x) and it

called the probability density function (pdf) of X.
f (x) is pdf if satisfies these rules:
1 f (x) ⩾ 0.
R∞
2 −∞ f (x)dx = 1 ( The total area under f (x) is 1).
3 P(a ≤ X ≤ b) is area under the curve between a and b.

For any continuous random variable

P (a ≤ x ≤ b) = P (a < x < b) = P (a < x ≤ b) = P (a ≤ x < b)

STATS 2B03 (Lecture 3) 23 / 30 2

Normal Distribution (Gaussian Distribution)

Symmetric, bell shaped distribution. pdf of Normal

A random variable X take any value

between −∞ and ∞.
A continuous random variable X is said
to have a Normal distribution if it has
the following pdf
1 2 2
f (x) = √ e−(x−µ) /2σ
σ 2π
Where −∞ < x < ∞, µ ∈ R, σ > 0.
We denoted this by X ∼ N (µ, σ 2 ).

STATS 2B03 (Lecture 3) 24 / 30 2

Normal Distribution (Gaussian Distribution)

µ and σ 2 the Normal Distribution

mean and
N ~ (0, 1)
variance affect N ~ (2, 2)
N ~ (0, 0.5)

0.8
the location and N ~ (−2, 2)

shape of the pdf,

0.6
respectively. f(x)

0.4
0.2
0.0

−6 −4 −2 0 2 4 6

STATS 2B03 (Lecture 3) 25 / 30 2

Standard Normal Distribution

When µ = 0 and σ 2 = 1, it called Standard Normal distribution.

Standard normal random variable denoted by Z.

Standard normal pdf denoted by ϕ(z).

1 1 2
ϕ(z) = √ e− 2 z
2π
where −∞ < z < ∞, µ ∈ R.

We denoted this by Z ∼ N (0, 1)

Probabilities: If Z is a standard normal random variable then

P (a ≤ Z ≤ b) the area under standard normal curve between a and b

STATS 2B03 (Lecture 3) 26 / 30 2

Areas Under the Standard Normal Curve

Table A-2 (pages 626 and 627) gives the area under the standard
normal curve to the left of z

Example: Find the area under the standard normal curve

1 To the left of z = 1.83

2 To the right of z = −2.17

3 Between z = 0.87 and z = 3.15

STATS 2B03 (Lecture 3) 27 / 30 2

Example

Find:

1 P (−1.25 < z ≤ 2.46)

2 Find z1 such that P (z1 < z < 1.87) = 0.8340

STATS 2B03 (Lecture 3) 28 / 30 2

Example

Let Z ∼ N (0, 1). Find z0

1 P (Z ≤ z0 ) = 0.7123

2 P (Z < z0 ) = 0.3405

3 P (Z ≥ z0 ) = 0.1515

STATS 2B03 (Lecture 3) 29 / 30 2

z−Score

Notation zα is a z−score with the property that the area under the
standard normal curve to the right of zα is equal to α.

In some contexts zα is called a critical value.

Example: Find zα/2 if

1 α = 0.1

2 α = 0.05

3 α = 0.01

STATS 2B03 (Lecture 3) 30 / 30 3