S2 Chapter 1: Binomial

Distribution
Dr J Frost ([email protected])
www.drfrostmaths.com

Last modified: 30th August 2015

Probability Distributions
In S1 we saw that a random variable has an associated probability distribution.
This had two components:

1 Outcomes
expressed as a set

2 A probability distribution
which maps outcomes to probabilities

Description Outcomes Probability Function

Throwing a fair die. {1,2,3,4,5,6} 1
? 𝑃 𝑋 = 𝑥 ?= 6 for all 𝑥

Number of heads seen 1

{0,1,2} 𝑥 = 0,2
after throwing a coin Note that we said on the left 𝑃 𝑋=𝑥 = 4
twice. ?
that the count is the outcome,
1 ? 𝑥=1
not what each coin was! 2
Off-The-Shelf Probability Distributions
We can define the probability function with whatever outcomes and probabilities we like.

However, there are certain ‘off-the-shelf’ probability distributions where the probability
function is already written for us and formulae are given for 𝐸[𝑋] and 𝑉𝑎𝑟[𝑋] – we have
some limited control over them by setting the values of some variables known as
‘parameters’.

We encountered two such ‘off-the-shelf’ distributions in S1:

These are all based on the parameters we set.

Description Name Params Outcomes Prob Func 𝑬𝑿 𝑽𝒂𝒓 𝑿
A Bell-shaped Normal Mean 𝜇 ℝ 1 −
𝑥−𝜇 2 𝜇 𝜎2
𝑝 𝑥 = 𝑒 2𝜎 2
distribution around Distribution Variance 𝜎 2 (Any real value)
some known mean
with a known variance.
?
𝑁 𝜇, 𝜎 2 ? ? ?
𝜎 2𝜋
Oddly, this is in your S1 formula
sheet, but you never used it!
? ?
Each discrete outcome Discrete Number of {1,2, … , 𝑛} 1 𝑛+1 𝑛2 − 1
𝑝 𝑥 = 𝑓𝑜𝑟 𝑎𝑙𝑙 𝑥
is equally likely to Uniform outcomes 𝑛
happen. ?
Distribution ? ? 𝑛
? ?2
?12

(No need to copy this – We will keep expanding

this table and coming back to it)
 Off-The-Shelf Probability Distributions
In Chapters 1, 2, and 4, we will gradually explore three new ‘off-the-shelf’ distributions.
Don’t worry yet!
These are all based on the parameters we set.
Description Name Params Outcomes Prob Func 𝑬𝑿 𝑽𝒂𝒓 𝑿
A Bell-shaped Normal Mean 𝜇 ℝ 1 −
𝑥−𝜇 2 𝜇 𝜎2
𝑝 𝑥 = 𝑒 2𝜎 2
distribution around Distribution Variance 𝜎 2 (Any real value) 𝜎 2𝜋
some known mean with 𝐍 𝛍, 𝛔𝟐 Oddly, this is in your S1 formula
a known variance. sheet, but you never used it!

Each discrete outcome is Discrete Number of {1,2, … , 𝑛} 1 𝑛+1 𝑛2 − 1

𝑝 𝑥 = 𝑓𝑜𝑟 𝑎𝑙𝑙 𝑥
equally likely to happen. Uniform outcomes 𝑛 𝑛 2 12
Distribution

We count the number of Binomial Number of trials 𝑛 0, 1, … , 𝑛 𝑛 𝑥 𝑥 𝑛𝑝 𝑛𝑝 1 − 𝑝

𝑝 𝑥 = 𝑝 1−𝑝
‘successes’ after a Distribution 𝑥
number of trials, each 𝑩 𝒏, 𝒑 Probability of
with two outcomes success in each
(‘success’ and ‘failure’). trial 𝑝
e.g. Number of heads
after 10 throws of an
unfair coin.
Counting the number of Poisson Average rate 𝜆 ℕ −𝜆
𝜆𝑥 𝜆 𝜆
𝑝 𝑥 =𝑒
events which occur Distribution (including 0) 𝑥!
within a fixed time, 𝑷 𝝀
given some known rate.
Outcome could be any Continuous Upper and lower 𝑎, 𝑏 1 𝑎+𝑏 2
𝑏−𝑎
𝑝 𝑥 =
number within some Uniform limit 𝑎 and 𝑏 𝑏−𝑎 2 12
continuous range 𝑎 to 𝑏 Distribution
𝑼 𝒂, 𝒃
Back in 2010 I was on holiday in Hawaii and visited the family of a friend. We noticed that
at the dinner table that out of the 8 of us, 6 of us were left-handed (including myself).
One of them asked, “The chances of that must be very low”.
I saw that as a challenge.

We’ll solve this problem in a sec…

S2 – Chapter 2 – Binomial Distribution

Lesson 1: Introduction to Combinatorics

Factorial and Choose Function
Q1 How many ways are there of arranging 5 different coloured beads in a line?

There are 5 choices for the first beads. 4 for the second, and so on.
So 5 × 4 × 3 × 2 × 1 = 5! = 120 possibilities.
? the number of ways of
𝒏! is the factorial function, and gives us
arranging 𝒏 distinguishable objects.
Factorial and Choose Function
How many ways are there of arranging 2 red beads and 3 blue beads in a line?
Q2
We can’t distinguish each of the red beads, nor the blue beads.

These 2 arrangements are actually the same!

Click for
Bromanimation
𝑅1 𝐵1 𝐵2 𝑅2 𝐵3
Answer

#3: There are similarly 3!

Ways of arranging the blue
#1: As before there are 5!
beads, so we need to divide
ways of arranging the 5
by 3!
beads.
5!
#4: So there are = 10 ways of arranging the beads.
3!2!
#2: However, since there Another way of thinking about it is that we’re
are 2! ways of arranging the “choosing” 2 of the 5 slots in the line for the red beads
red beads, each 2! to go in.
possibilities represents the 𝟓
same arrangement, so we We can write this as said “5 choose 2”.
𝟐
divide by 2! to account for 𝑛 𝑛!
duplicates. = is the number of ways of
𝑟 𝑟! 𝑛−𝑟 !
choosing 𝑟 items from 𝑛 where the order of
the items in the choice don’t matter.
Test Your Understanding
T1 How many ways of arranging the T4 I throw a coin 8 times. How many
letters in the word TELEVISION? possibilities are there in which I threw
There are 10! ways of arranging 10 4 Heads?
distinct things.
But the 2! ways of arranging the E’s Example sequences: HHTTHTTH,
lead to the same word. HHHHTTTT, …
And the 2! Ways of arranging the I’s By the same logic as before, we are
?
lead to the same word. choosing 4 of the 8?positions in the
𝟖
sequences to put ‘H’ in, so = 𝟕𝟎
𝟏𝟎! 𝟒
= 𝟗𝟎𝟕𝟐𝟎𝟎 possibilities.
𝟐! 𝟐!
If you have 8 people at a dinner table, how
T2 How many ways of arranging 3 red
T5 many ways are there for 6 of them to be
balls, and 4 green balls in a line? left-handed?
𝟕!
? = 𝟑𝟓
𝟑! 𝟒! We’re counting the possible choices of 6
people out of 8.
𝟖 ?
T3 How many ways of arranging 4 red That’s = 𝟐𝟖 possibilities.
𝟔
balls, 3 green balls and 2 blue balls in
a line?
𝟗!
𝟒! 𝟑! 𝟐!?
= 𝟏𝟐𝟔𝟎
 Exercise 1X (Not in textbook)

5 I’m a waiter serving a table of 10. I have 7 fish dishes

1 How many ways are there of arranging 10 (each the same) and 3 veg dishes (each the same).
red balls and 5 blue balls in a line. How many ways can meals be served to the table?
𝟏𝟓! We can choose 3 people to give the fish dishes to out
𝟏𝟎! 𝟓! ?
= 𝟑𝟎𝟎𝟑
𝟏𝟎 ?
of the 10. = 𝟏𝟐𝟎
𝟑
2 How many ways are there of arranging
the letters in the word BANANA? I throw a fair dice 20 times. How many possibilities are
𝟔! 6 there in which I saw a 1 eight times?
𝟑! 𝟐!?= 𝟔𝟎 𝟐𝟎
𝟖 ?
= 𝟏𝟐𝟓𝟗𝟕𝟎

3 How many sequences can we make

consisting of 3 A’s, 4 B’s, and 5 C’s (e.g. 7 I have 10 different pieces of fruit to pick from. I pick 3,
ABCCBAACBCC) and put them in a line. How many possible lines of
𝟏𝟐! fruit are there?
𝟑! 𝟒! 𝟓! ?
= 𝟐𝟕𝟕𝟐𝟎 𝟏𝟎
𝟑 ?
× 𝟑! = 𝟕𝟐𝟎

4 Given that in the lottery you pick 6 Hard: How many possible lottery tickets are there
distinct numbers from 1 to 49, what’s the 8
where exactly 3 of the 6 numbers match (given there
probability you win the lottery?
𝟒𝟗 were again 49 numbers to choose from)
There are = 𝟏𝟑𝟗𝟖𝟑𝟖𝟏𝟔 possible There are
𝟔
possible ways of choosing the 3 numbers
𝟔
?
tickets. So probability is
𝟏 𝟑
which matched. However there are
𝟒𝟑
ways of choosing
𝟏𝟑𝟗𝟖𝟑𝟖𝟏𝟔
? 𝟑
𝟔 𝟒𝟑
the 3 numbers which didn’t match. So × =
𝟑 𝟑
𝟐𝟒𝟔𝟖𝟐𝟎
Probability based Questions
A fair die is rolled 8 times. Find the probability of:
a) No sixes
Q b) Only 3 sixes
c) 4 twos and 4 sixes

a Probability of not rolling a six is 5

6

𝑃 𝑛𝑜 𝑠𝑖𝑥𝑒𝑠 =
5
8
?
6

b 3 were six and 5 weren’t six.

If ‘S’ means a six was thrown and ‘N’ means it wasn’t throw, one possibility is
1 3 5 5
SSNNNSNN, which has probability
6 6
But there’s
8
=
8! ?
ways of arranging 3 S’s and 5 N’s. So:
3 3!5!
3 5
8 1 5
𝑃 3 𝑠𝑖𝑥𝑒𝑠 = × = 0.104
3 6 6
4 4
c 1 1 8
𝑃 4 𝑡𝑤𝑜𝑠 𝑎𝑛𝑑 4 𝑠𝑖𝑥𝑒𝑠 = × ×
6 6 3
= 0.0000417 ?
S2 – Chapter 2 – Binomial Distribution

Lesson 2: Binomial Distribution

Back in 2010 I was on holiday in Hawaii and visited the family of a friend. We noticed that
at the dinner table that out of the 8 of us, 6 of us were left-handed (including myself).
One of them asked, “The chances of that must be very low”.
I saw that as a challenge.

Consider the 8 people in a line. Suppose 10% of

the population is left-handed. Can you now work
out the probability of 6 being left-handed now?

One possibility say is LLRLLRLL.

The probability of this particular possibility is 𝟎. 𝟏 ×
𝟎. 𝟏 × 𝟎. 𝟗 × 𝟎. 𝟏 × ⋯ = 𝟎. 𝟏𝟔 × 𝟎. 𝟗𝟐

𝟖
But there are = 𝟐𝟖 possible ways in which 6
𝟔 ?
people in the room could be left handed.

𝟖
So probability is × 𝟎. 𝟏𝟔 × 𝟎. 𝟗𝟐 = 𝟎. 𝟎𝟎𝟎𝟐𝟐𝟔𝟖 ≈
𝟔
𝟏 𝒊𝒏 𝟒𝟒 𝟎𝟎𝟎
 Test Your Understanding
I throw a unfair coin 10 times, where the probability of heads is 𝑝. What’s the
probability of throwing 3 heads?
Q Let 𝑋 be a random variable which counts the number of heads seen in 10 throws.
Determine the probability distribution.
For convenience you can use 𝑞 as the probability of tails (where 𝑞 = 1 − 𝑝)

𝒙 𝟎 𝟏 𝟐 𝟑 𝟒 𝟓 𝟔 𝟕 𝟖 𝟗 𝟏𝟎

𝑃 𝑋=𝑥 𝑞10 10
𝑝𝑞 9
10 2 8 10 3 7 10 4 6 10 5 5 10 6 4 10 7 3 10 8 2 10 9 1 𝑝10
? ?
1 ?
2
𝑝 𝑞
?
3
𝑝 𝑞
?
4
𝑝 𝑞
?
5
𝑝 𝑞
?
6
𝑝 𝑞
?
7
𝑝 𝑞
?
8
𝑝 𝑞
9 ?
𝑝 𝑞
?

Notice that each of the terms form the Binomial Expansion of

𝑝 + 𝑞 10 (as per C2). Since 𝑝 + 𝑞 = 1, 𝑝 + 𝑞 10 = 1, which shows
that the probabilities add to 1 (as we’d hope!)

For this reason we call this type of distribution a Binomial Distribution.

Binomial Distribution
e.g. We threw a coin 𝑛 times.
! We have a Binomial Distribution when we have:

1. a FIXED number 𝑛 trials.

So that we’re allowed to multiply
2. The trials are INDEPENDENT. the probabilities together from
3. A CONSTANT probability 𝑝 of success (in each trial) each trial

4. TWO outcomes in each trial, “success” and “failure”. We’re counting heads, so throwing
a heads is the “success”.

If these conditions are met, a random variable 𝑋 whose outcome is the

number of successes is Binomially distributed:

𝑋~𝐵 𝑛, 𝑝

𝑛 𝑥 𝑛−𝑥
𝑃 𝑋=𝑥 = 𝑝 1−𝑝
𝑥

𝑥 is number of successes. If 𝑛 trials and 𝑥 successes,

must be 𝑛 − 𝑥 failures.

FICT
Quickfire Questions
Show the calculation required to find the indicated probability given the distribution.

𝟏𝟎
𝑋~𝐵 10,0.3 𝑷 𝑿=𝟒 = 𝟎. 𝟑𝟒? 𝟎. 𝟕𝟔
𝟒

𝟏𝟎
𝑋~𝐵 10,0.2 𝑷 𝑿=𝟑 = 𝟎. 𝟐𝟑? 𝟎. 𝟖𝟕
𝟑

𝟓
𝑋~𝐵 5,0.1 𝑷 𝑿=𝟐 = 𝟎. 𝟏𝟐 ?𝟎. 𝟗𝟑
𝟐

𝑋~𝐵 20,0.45 𝑷 𝑿 = 𝟐𝟎 = 𝟎. 𝟒𝟓𝟐𝟎 ?

𝑋~𝐵 20,0.45 𝑷 𝑿 = 𝟎 = 𝟎. 𝟓𝟓𝟐𝟎 ?

 Is it Binomially Distributed?
Is a Binomial Distribution appropriate as a model? The number of red
balls selected
when 3 balls are
Some number out drawn from bag of
of 8 people being Number of throws on Number of girls in 15 white and 5 red
left-handed die until 6 obtained family of 4 children balls.


1. We have a fixed No, not fixed. This is
number 𝑛 trials.  known as a ‘Geometric
Distribution’ (which we

won’t cover)
2. Each trial has two
possibilities,
“success” and    

“failure”.
?
Usually. But in my
? ? ?
3. The trials are Technically the probability of Only if balls drawn

story, genetics has having a girl increases if you
independent. previously had a girl, and vice with replacement.
an influence on
versa. But the probability is
handedness. still close to 0.5, so Binomial
Distribution is appropriate.

Only if balls drawn

4. There is a fixed
probability 𝑝 of    with replacement,
𝑝=
1
success in each trial. 4
Test Your Understanding
1
Q1 𝑋~𝐵 12,
6
What is 𝑃(𝑋 = 2)?
𝟐 𝟏𝟎
𝑷 𝑿=𝟐 =
𝟏𝟐
𝟐
?𝟏𝟔 𝟓
𝟔
= 𝟎. 𝟐𝟗𝟔
What is 𝑃(𝑋 ≤ 1)?
𝑷 𝑿≤𝟏 =𝑷 𝑿=𝟎 +𝑷 𝑿=𝟏
𝟏𝟐 𝟏 𝟏𝟏
𝟓 𝟏𝟐 𝟏 𝟓
= +?
𝟔 𝟏 𝟔 𝟔
= 𝟎. 𝟑𝟖𝟏

Q2 I have a bag of 2 red and 8 white balls. 𝑋 represents the number of

red balls I chose after 5 selections (with replacement).

a How is 𝑋 distributed?
𝑿~𝑩 ?𝟓, 𝟎. 𝟐
b Determine the probability that I chose 3 red balls.
𝟓
𝑷 𝑿=𝟑 = 𝟐𝟑 𝟎. 𝟖𝟐 = 𝟎. 𝟎𝟓𝟏𝟐
𝟎.?
𝟑
(If you get these quickly, go on to Exercise 1B)
Exercise 1B
1 A student suggests using a binomial distribution
1 The random variable 𝑋~𝐵 8, 3 5 to model the following situations. Give a
Find description of the random variable, state any
assumptions that must be made and give
𝑃 𝑋 = 2 = 𝟎. 𝟐𝟕𝟑
possible values for 𝑛 and 𝑝.
?
𝑃 𝑋 = 5 = 𝟎. 𝟎𝟔𝟖𝟑
𝑃 𝑋 ≤ 1 = 𝟎. 𝟏𝟗𝟓 a A sample of 20 bolts is checked for defects from a
large batch. The production process should
2 produce 1% of defective bolts.
3 The random variable 𝑇~𝐵 15, 𝑿~𝑩 𝟐𝟎, 𝟎. 𝟎𝟏 assuming bolts being defective
Find
3
?
are independent from each other.

?
𝑃 𝑇 = 5 = 𝟎. 𝟎𝟎𝟔𝟕𝟎 b Some traffic lights have three phases: stop 48% of
?
𝑃 𝑇 = 10 = 𝟎. 𝟐𝟏𝟒 the time, wait or get ready 4% of the time and go
?
𝑃 3 ≤ 𝑇 ≤ 4 = 𝟎. 𝟎𝟎𝟏𝟕𝟖 48% of the time. Assuming that you only cross a
traffic light when it is in the go position, model
the number of times that you have to wait or
4 A balloon manufacturer claims that 95% stop on a journey passing through 6 sets of traffic
of his balloons will not burst when blown lights.
𝑿~𝑩 𝟔, 𝟎. 𝟓𝟐 assuming lights operate
up. You have 20 balloons. independently. ?
What is the probability that none of
them burst? c When Stephanie plays tennis with Tim on
𝟎. 𝟗𝟓𝟐𝟎 = 𝟎.? 𝟑𝟓𝟖 𝒕𝒐 𝟑𝒔𝒇 average one in eight of her serves is an ‘ace’. How
many ‘aces’ does Stephanie serve in the next 30
What is the probability exactly 2 burst? serves against Tim?
= 𝟎. 𝟏𝟖𝟗 ? 𝟏
𝑿~𝑩 𝟑𝟎, assuming serves are independent
𝟖
?
and probability of an ace is constant.
 Overview So Far

These are all based on the parameters we set.

Description Name Params Outcomes Prob Func 𝑬𝑿 𝑽𝒂𝒓 𝑿
We count the number of Binomial Number of trials 𝑛 0, 1, … , 𝑛 𝑝 𝑥 =
‘successes’ after a
number of trials, each
Distribution
𝑩 𝒏, 𝒑 Probability of
𝑛 𝑥
𝑥
𝑝 1−𝑝 𝑛−𝑥
? ?
with two outcomes
(‘success’ and ‘failure’). ?
success in each
trial 𝑝 ? ?
e.g. Number of heads
after 10 throws of an
unfair coin.

Still to cover:
• The Cumulative Distribution Function 𝐹(𝑥)
• Calculating 𝐸 𝑋 and 𝑉𝑎𝑟 𝑋
S2 – Chapter 2 – Binomial Distribution

Lesson 3: Cumulative Distribution Function

Cumulative Distribution Function
In S1, the Cumulative Distribution Function is defined as:
𝑭 𝒙 = 𝑷 𝑿? ≤𝒙
and means: “the probability of the outcome ? being up to a particular value”.
We saw in S1 that for the Normal Distribution, we had to look up 𝐹 𝑥 values in a z-table,
because they were not possible to calculate with a simple formula.
There’s a similar problem here: To calculate 𝐹 10 = 𝑃 𝑋 = 1 + 𝑃 𝑋 = 2 + ⋯ +
𝑃 𝑋 = 10 would be computationally expensive, as the summation doesn’t simplify and the
factorial function is an expensive operation (your calculator takes roughly half a second to
compute 50!).

We therefore use tables (From Page 135 in the newer S2 Edexcel textbooks).

The random variable 𝑋~𝐵(20,0.4). Find:

Look up 𝑛 = 20, 𝑝 = 0.4, 𝑥 = 7
?
𝑃 𝑋 ≤ 7 = 0.4159
𝑃 𝑋 < 6 = 𝑃 𝑋 ≤ 5 ?= 0.1256
? = 0.0016
𝑃 𝑋 ≥ 15 = 1 − 𝑃 𝑋 ≤ 14
 Cumulative Distribution Function
We can also use the table to find when 𝑋 = 𝑥 or 𝑎 ≤ 𝑋 ≤ 𝑏.

But you would never need

to do this in an exam –
Given that 𝑋~𝐵 25,0.25 even for the biggest value
of 𝑛 in the table, your
𝑃 𝑋 =6 =𝑷 𝑿≤𝟔 −𝑷 𝑿≤𝟓 calculator could calculate
? this directly!
= 𝟎. 𝟏𝟖𝟐𝟖
𝑃 𝑋 > 20 = 𝟏 − 𝑷 𝑿?≤ 𝟐𝟎
𝑃 6 < 𝑋 ≤ 10 = 𝑷 𝑿 ≤ 𝟏𝟎 ?− 𝑷(𝑿 ≤ 𝟔) Be careful about < vs ≤

Quickfire Questions

Write the following in terms of cumulative probabilities, e.g. 𝑃 𝑋 < 7 = 𝑃 𝑋 ≤ 6

𝑃 𝑋 < 5 = 𝑷 𝑿?≤𝟒 ?
𝑃 10 ≤ 𝑋 ≤ 20 = 𝑷 𝑿 ≤ 𝟐𝟎 − 𝑷 𝑿 ≤ 𝟗
𝑃 𝑋 ≥ 7 = 𝟏 − 𝑷 ?𝑿 ≤ 𝟔 ?
𝑃 𝑋 = 100 = 𝑷 𝑿 ≤ 𝟏𝟎𝟎 − 𝑷 𝑿 ≤ 𝟗𝟗
𝑃 𝑋 > 7 = 𝟏 − 𝑷 ?𝑿 ≤ 𝟕 ?
𝑃 20 < 𝑋 < 30 = 𝑷 𝑿 ≤ 𝟐𝟗 − 𝑷 𝑿 ≤ 𝟐𝟎
𝑃 10 ≤ 𝑋 < 20 = 𝑷 𝑿 ≤ 𝟏𝟗 ?− 𝑷 𝑿 ≤ 𝟗 ?
“at least 30” = 𝑷 𝑿 ≥ 𝟑𝟎 = 𝟏 − 𝑷 𝑿 ≤ 𝟐𝟗
?
“greater than 30” = 𝑷 𝑿 > 𝟑𝟎 = 𝟏 − 𝑷 𝑿 ≤ 𝟑𝟎
Test Your Understanding
An awkward Tiffin boy ventures into Tiffin Girls. He asks 20 girls out on the date. The
probability that each girl says yes is 0.3.
Determine the probability that he will end up with:
a) Less than 6 girls on his next date.
b) At least 9 girls on his next date.
Q
The boy considers the evening a success if he dated at least 9 girls that evening.
He repeats this process across 5 evenings.
c) Calculate the probability that he had at least 4 successful evenings.
(Note: You won’t be able to use your table for (c) as 𝑝 is not a nice round number – calculate prob directly)

a 𝑋~𝐵 20, 0.3 𝑋 is the number of girls dated in an evening.

𝑃 𝑋<6 =𝑃 𝑋≤5 ?
= 𝟎. 𝟒𝟏𝟔𝟒
b 𝑃 𝑋 ≥9 =1−𝑃 𝑋 ≤8
= 𝟏 − 𝟎. 𝟖𝟖𝟔𝟕 ?
= 𝟎. 𝟏𝟏𝟑𝟑
c 𝑌~𝐵 5, 0.1133 𝑌 is the number of successful evenings.
𝑷 𝑿≥𝟒 =𝑷 𝑿=𝟒 +𝑷 𝑿=𝟓
=
𝟓
𝟎. 𝟏𝟏𝟑𝟑𝟒 × 𝟎. 𝟖𝟖𝟔𝟕 + ? 𝟎. 𝟏𝟏𝟑𝟑𝟓
𝟒
= 𝟎. 𝟎𝟎𝟎𝟕𝟒𝟗
Exercise 1C
1 𝑋~𝐵 9, 0.2 8 In a town, 30% of the residents listen to the local
𝑃 𝑋 ≤ 4 = 𝟎. 𝟗𝟖𝟎𝟒 radio. Ten residents are chosen at random.
𝑃 𝑋 < 3 = 𝟎. 𝟕𝟑𝟖𝟐 ? State the distribution for the random variable.
a
𝑃 𝑋 ≥ 2 = 𝟎. 𝟓𝟔𝟑𝟖
? ?
𝑿~𝑩(𝟏𝟎, 𝟎. 𝟑𝟎)
𝑃 𝑋 = 1 = 𝟎. 𝟑𝟎𝟐𝟎 Find the probability that at least half of these 10
b
residents listen to local radio.
3 𝑋~𝐵(40,0.45). Find ?
𝟎. 𝟏𝟓𝟎𝟑
𝑃 𝑋 < 20 = 𝟎. 𝟔𝟖𝟒𝟒
𝑃 𝑋 > 16 = 𝟎. 𝟔𝟖𝟏𝟓 ?
𝑃 11 ≤ 𝑋 ≤ 15 = 𝟎. 𝟐𝟎𝟔𝟖 Further Question (not in book)
𝑃 10 < 𝑋 < 17 = 𝟎. 𝟑𝟏𝟏𝟏 ?
Light bulbs come in packs of 6. The probability of any given
lightbulb being faulty is 0.05.
What’s the probability that more than 3 bulbs are faulty.
a 𝑿~𝑩 𝟔, 𝟎. 𝟎𝟓
𝑷 𝑿>𝟑 =𝟏−𝑷 𝑿≤𝟑
= 𝟎. 𝟎𝟎𝟎𝟏
?
b A pack can only be sold if it contains no faulty bulbs.
Calculate the probability that out of 10 packs, exactly 8 can
be sold.
𝑷 𝑿 = 𝟎 = 𝟎. 𝟗𝟓𝟔 = 𝟎. 𝟕𝟑𝟓𝟏
𝒀~𝑩 𝟏𝟎, 𝟎. 𝟕𝟑𝟓𝟏 𝒀 is number of non-faulty packs.
𝑷 𝒀=𝟖 =
𝟏𝟎 ?
𝟎. 𝟕𝟑𝟓𝟏𝟖 𝟎. 𝟐𝟔𝟒𝟗𝟐 = 𝟎. 𝟐𝟔𝟗𝟐
𝟖

You can start on your homework questions if you’ve done all of these (ask for sheet).
S2 – Chapter 2 – Binomial Distribution

Lesson 4: Probability Ranges

Dealing with Probability Ranges
A spinner is designed so that probability it lands on red is 0.3. Jane has 12 spins.
a) Find the probability that Jane obtains at least 5 reds.
Jane decides to use this spinner for a class competition. She wants the probability of
Q
winning a prize to be < 0.05. Each member of the class will have 12 spins and the
number of reds will be recorded.
b) Find how many reds are needed to win the prize.

𝑃 𝑋 ≥ 5 = 1 − 𝑃 𝑋 ≤ 4 = 0.2763 STEP 1: Represent the

sentence using
𝑃 𝑋 ≥ 𝑟 < 0.05 probability.

1 − 𝑃 𝑋 ≤ 𝑟 − 1 < 0.05 STEP 2: Ensure LHS involves ≤ inside

𝑃 𝑋 ≤ 𝑟 − 1 > 0.95 probability.
𝑟−1=6
𝒓=𝟕
? STEP 3: Rearrange.

STEP 4: Use table backwards to find value

corresponding to closest probability.

Note that the textbook does this in a less methodical way: but the method above is
what you find in exam mark schemes, so ignore at your peril.
Test Your Understanding
At Camford University, students have 20 exams at the end of the year. All students
pass each individual exam with probability 0.45. Students are only allowed to continue
Q into the next year if they pass some minimum of exams out of the 20.
What do the university administrators set this minimum number such that the
probability of continuing to next year is at least 90%?

𝑋~𝐵 20, 0.45

𝑃 𝑋 ≥ 𝑘 ≥ 0.9
1 − 𝑃 𝑋 ≤ 𝑘 − 1 ≥ 0.9
𝑃 𝑋 ≤ 𝑘 − 1 ≤ 0.1
𝑘−1=5 ?
𝒌=𝟔
This is exactly what
you should write.
Exercise 1C (again)
𝑋~𝐵 50,0.40 . Find B Each question in a test has a 50%
7
The largest value of 𝑘 such that chance of being answered
𝑃 𝑋 ≤ 𝑘 < 0.05 𝒌 = 𝟏𝟑 ? correctly. What minimum pass
The smallest number 𝑟 such that
mark out of 10 is required such
𝑃 𝑋 > 𝑟 < 0.01 ?
𝒓 = 𝟐𝟖
that at least 75% of students will
8 𝑋~𝐵 50,0.10 . Find pass?
The largest value of 𝑘 such that 𝑿~𝑩 𝟏𝟎, 𝟎. 𝟓
𝑃 𝑋 < 𝑘 < 0.02 𝒌=𝟏 ? 𝑷 𝑿 ≥ 𝒌 ≥ 𝟎. 𝟕𝟓
The smallest number 𝑟 such that 𝑷 𝑿 ≤ 𝒌 − 𝟏 ≤ 𝟎. 𝟐𝟓
𝑃 𝑋 > 𝑟 < 0.01 ?
𝒓 = 𝟏𝟎 𝒌 − 𝟏 = 𝟑?
𝑃 𝑘≤𝑋≤𝑟 ?
𝟎. 𝟗𝟖𝟎𝟏 𝒌=𝟒

Further Questions (not in book)

A 𝑋~𝐵 25, 0.2
Find the smallest value of 𝑘 such that
𝑃 𝑋 ≥ 𝑘 < 0.05
𝑷 𝑿 ≤ 𝒌 − 𝟏 > 𝟎. 𝟗𝟓
𝒌−𝟏=𝟖 ? → 𝒌=𝟗
S2 – Chapter 2 – Binomial Distribution

Lesson 5: E[X], Var[X], and tables with 𝑝 > 0.5

Mean 𝐸 𝑋 and Variance 𝑉𝑎𝑟 𝑋
I want to know amongst 200 people the probability of a certain number being
left handed, where the probability any given person is left handed is 0.1.

𝑋~𝐵 200, 0.1 gives a binomial distribution over the possible numbers of left-
handed people.

Based on what you learnt in Year 8/9, how many people would you expect on
average to be left-handed?

𝟐𝟎𝟎 × 𝟎. 𝟏 = 𝟐𝟎
i.e. We multiplied the number of trials 𝒏 and the probability of success 𝒑 to
work out the mean number of successes.
?
The variance is much harder to work out and we won’t give the proof here.

!If 𝑋~𝐵 𝑛, 𝑝
𝐸 𝑋 = 𝜇 = 𝑛𝑝 Bro Tip: It’s perhaps easier to
𝑉𝑎𝑟 𝑋 = 𝜎 2 = 𝑛𝑝 1 − 𝑝 remember variance as 𝑛𝑝𝑞
(where 𝑞 = 1 − 𝑝 is the
probability of failure, although
these are given the formula book)
Examples
A fair, 4-sided die has the numbers 1, 2, 3, 4 on its faces. The die is rolled 20 times. The random
variable 𝑋 represents the number of 4s obtained.

Find the mean and variance of 𝑋.

Q 𝑿~𝑩 𝟐𝟎, 𝟎. 𝟐𝟓
𝑬 𝑿 = 𝟐𝟎 × 𝟎. 𝟐𝟓 = 𝟓 ?
𝑽𝒂𝒓 𝑿 = 𝟐𝟎 × 𝟎. 𝟐𝟓 × 𝟎. 𝟕𝟓 = 𝟑. 𝟕𝟓
Find 𝑃 𝑋 < 𝜇 − 𝜎
𝝈 = 𝟑. 𝟕𝟓 = 𝟏. 𝟗𝟑𝟔 …
?
𝑷 𝑿 < 𝝁 − 𝝈 = 𝑷 𝑿 < 𝟑. 𝟎𝟔. . = 𝑷 𝑿 ≤ 𝟑 = 𝟎. 𝟐𝟐𝟓𝟐

David believes that 35% of people in a certain town will vote for him in the next election and he
commissions a survey. Find the minimum number of people the survey should ask to have a
Q mean number of more than 100 voting for David.
𝟎. 𝟑𝟓𝒏 = 𝟏𝟎𝟎 𝒏 = 𝟐𝟖𝟓. 𝟕
?
So 𝟐𝟖𝟔 people should be asked.

An examiner is trying to design a multiple choice test. For students answering the test at
random, he requires that the mean score on the test should be 20 and standard deviation at
least 4. Find how many choices each question should have and the number of questions there
should be. Number of choices is fixed across questions and show be as small as possible.
Q 𝒏𝒑 = 𝟐𝟎
𝒏𝒑 𝟏 − 𝒑 ≥ 𝟒 → 𝟐𝟎 𝟏 − 𝒑 ≥ 𝟒 → 𝒑 ≤ 𝟎. 𝟐
So there should be 5 choices per question. ?
𝒏𝒑 = 𝟐𝟎 → 𝟎. 𝟐𝒏 = 𝟐𝟎 → 𝒏 = 𝟏𝟎𝟎
Exercise 1D
Using tables when 𝑝 > 0.5
Just as the z-table for the Normal Distribution only had probabilities greater than 0.5,
the Cumulative Distribution Function tables only have values for 𝑝 less than 0.5!
We similarly have to do some manipulation when 𝑝 > 0.5.

In Joe’s café 70% of customers buy a cup of tea.

a) In a random sample of 20 customers find the probability that more than
15 buy a cup of tea.

Let 𝑿 = the number out of 20 who buy a cup of tea.

?
𝑿~𝑩 𝟐𝟎, 𝟎. 𝟕
Let 𝒀 = the number of customers who do not buy a cup of tea.
Hint: Perhaps rethink the problem in terms of who doesn’t buy tea.
𝒀~𝑩 𝟐𝟎, 𝟎. 𝟑
𝑷 𝑿 > 𝟏𝟓 = 𝑷 𝒀 ≤ 𝟒 = 𝟎. 𝟐𝟑𝟕𝟓

Notes: By focusing on the opposite, we can get our 𝑝 to be above 0.5. If

more than 15 people buy tea, at most 4 didn’t drink tea. So we can
instead find the probability of this.

Bro Tip: In general, if 𝑋 < 𝑎, then 𝑌 > 𝑛 − 𝑎

And if 𝑋 ≤ 𝑎, then 𝑌 ≥ 𝑛 − 𝑎
So we just subtract from the number of trials and preserve the
strictness/non-strictness of the inequality.
Quickfire Questions
Change from the count of successes 𝑋 to the count of failures 𝑌,
ensuring you ends up with ≤.
Given 100 trials:

𝑃 ≥ 60 = 𝑷 𝒀 ≤? 𝟒𝟎
𝑋
𝑃 < 30 = 𝑃 𝑌 > 70 = 𝟏? − 𝑷 𝒀 ≤ 𝟕𝟎
𝑋
𝑃 > 20 = 𝑃 𝑌 < 80 =? 𝑷 𝒀 ≤ 𝟕𝟗
𝑋
𝑃 𝑋
≥ 30 = 𝑷 𝒀 ≤ 𝟕𝟎 ?
𝑃 𝑋
≤ 25 = 𝑃 𝑌 ≥ 75 = 𝟏? − 𝑷 𝒀 ≤ 𝟕𝟒
𝑃 𝑋
< 30 ≤ 0.2
→ 𝑷 𝒀 > 𝟕𝟎 ≤ 𝟎. 𝟐
𝑷 𝒀 ≤ 𝟕𝟎 ≥ 𝟎.? 𝟖
𝑃 𝑋 ≤ 65 ≥ 0.55
→ 𝑷 𝒀 ≥ 𝟑𝟓 ≥ 𝟎. 𝟓𝟓
𝑷 𝒀 ≤ 𝟑𝟒 ≤ ?𝟎. 𝟒𝟓
Test Your Understanding
Historically, 75% of people who have visited KGS have been eaten by bears.
If 30 Tiffin Boys visit KGS, determine the probability that at least 20 of them will be eaten by bears.

Let 𝑿 be number of boys eaten by bears, and 𝒀 number who don’t.

𝑿~𝑩 𝟑𝟎, 𝟎. 𝟕𝟓
?
𝒀~𝑩(𝟑𝟎, 𝟎. 𝟐𝟓)
𝑷 𝑿 ≥ 𝟐𝟎 = 𝑷 𝒀 ≤ 𝟏𝟎 = 𝟎. 𝟖𝟗𝟒𝟑

S2 May 2012 Q8 (Part (b) was very challenging according to the Examiner’s Report!)

Note: they’re sneakily trying to

confuse you here because 𝑛 here
is not the number of trials!

𝑋~𝐵 50,0.6
𝑌~𝐵 50,0.4
𝑃 𝑋 < 𝑛 = 𝑃 𝑌 > 50 − 𝑛 ≥ 0.9
?
𝑃 𝑌 ≤ 50 − 𝑛 ≤ 0.1
50 − 𝑛 ≤ 15
𝑛 ≥ 35
𝑛 = 35
And more general probability questions…
In Joe’s café 70% of customers buy a cup of tea.
a) In a random sample of 20 customers find the probability that more than 15 buy a cup of
tea.
Let 𝑿 = the number out of 20 who buy a cup of tea.
𝑿~𝑩 𝟐𝟎, 𝟎. 𝟕
Let 𝒀 = the number of customers who do not buy a cup of tea.
𝒀~𝑩 𝟐𝟎, 𝟎. 𝟑
𝑷 𝑿 > 𝟏𝟓 = 𝑷 𝒀 ≤ 𝟒 = 𝟎. 𝟐𝟑𝟕𝟓

b) The proportion of customers who buy a chocolate muffin at Joe’s café is 0.35. Find the
probability that Joe sells the first chocolate muffin to his fifth customer.

The first four customers don’t buy a muffin, and the fifth does.
?
𝟎. 𝟔𝟓𝟒 × 𝟎. 𝟑𝟓 = 𝟎. 𝟎𝟔𝟐𝟓 (𝒕𝒐 𝟑𝒔𝒇)

c) Find the probability that in a random sample of 20 customers more than 15 buy a cup of
tea and a least 6 buy a chocolate muffin.

𝑷 𝑴≥𝟔 =𝟏−𝑷 𝑴≤𝟓

= 𝟎. 𝟕𝟓𝟒𝟔 ?
𝑷 𝑴 ≥ 𝟔 𝒂𝒏𝒅 𝑿 > 𝟏𝟓 = 𝟎. 𝟕𝟓𝟒𝟔 × 𝟎. 𝟐𝟑𝟕𝟓
= 𝟎. 𝟏𝟕𝟗 (𝟑 𝒔𝒇)
Exercises
The probability Pippin gets a stroke on any
1 Cats in general eat 65% of mice that they encounter. 4
Calculate the probability that for 10 mice Pippin given day is 0.7.
encounters in the next hour, she eats at least 6. a What is the minimum number of days she
𝑿~𝑩 𝟏𝟎, 𝟎. 𝟔𝟓 must be stroked over the next 20 so that
𝒀~𝑩 𝟏𝟎, 𝟎. 𝟑𝟓?
𝑷 𝑿 ≥ 𝟔 = 𝑷 𝒀 ≤ 𝟒 = 𝟎. 𝟕𝟓𝟏𝟓
there is at least a 95% chance I stroke her at
least this number of times?
𝑿~𝑩 𝟐𝟎, 𝟎. 𝟕
2 There is 80% chance that Pippin the cat wakes me 𝒀~𝑩 𝟐𝟎, 𝟎. 𝟑
up before 5.30am for food. Over the next week (7 𝑷 𝑿 ≥ 𝒌 ≥ 𝟎. 𝟗𝟓
days), what is the probability she wakes me up for ?
𝑷 𝒀 ≤ 𝟐𝟎 − 𝒌 ≥ 𝟎. 𝟗𝟓
𝟐𝟎 − 𝒌 = 𝟗
less than 4 of them?
𝒌 = 𝟏𝟏
𝑿~𝑩 𝟕, 𝟎. 𝟖
𝒀~𝑩 𝟕, 𝟎. 𝟐 b What is the probability that her first stroke
isn’t until the 5th day?
?
𝑷 𝑿<𝟒 =𝑷 𝒀>𝟑
𝟎. 𝟑𝟒 × 𝟎. 𝟕 = 𝟎. 𝟎𝟎𝟓𝟔𝟕
=𝟏−𝑷 𝒀≤𝟑
= 𝟎. 𝟎𝟑𝟑𝟑
?
5 90% of cats are black. Pippin has 50 kittens.
Forgetting genetics and that Pippin is
3 In my cupboard I have 3 boxes of Whiskers, 6 boxes
of Felix and 1 box of Waitrose Essentials Dry Food neutered, what’s the most number of
Mix, each box with a plentiful supply of meals. Over kittens such that I am at least 55% sure less
15 meals, what’s the probability Pippin eats Felix at than this number will be black.
most 5 times? 𝑿~𝑩 𝟓𝟎, 𝟎. 𝟗 𝒀~𝑩(𝟓𝟎 𝟎. 𝟏)
𝑿~𝑩 𝟏𝟓, 𝟎. 𝟔 𝒀~𝑩 𝟏𝟓, 𝟎. 𝟒 𝑷 𝑿 < 𝒌 ≥ 𝟎. 𝟓𝟓
𝑷 𝑿 ≤ 𝟓 = 𝑷 𝒀 ≥ 𝟏𝟎
= 𝟏−𝑷 𝒀≤𝟗 ? ?
𝑷 𝒀 > 𝟓𝟎 − 𝒌 ≥ 𝟎. 𝟓𝟓
𝑷 𝒀 ≤ 𝟓𝟎 − 𝒌 ≤ 𝟎. 𝟒𝟓
= 𝟎. 𝟎𝟑𝟑𝟖 𝟓𝟎 − 𝒌 = 𝟒 → 𝒌 = 𝟒𝟔
𝑛 or 𝑝 is unknown
The previous question was to establish an unknown 𝑥. But if the parameter 𝑛 or 𝑝
is unknown, we just have to use mathematical common sense (see part c).

S2 May 2013 Q7

? 𝟎. 𝟏
𝑿~𝑩 𝒏,

𝒀~𝑩 𝟏𝟎, 𝟎. 𝟏
𝑷 𝒀≥𝟒
?
=𝟏−𝑷 𝒀≤𝟑
= 𝟎. 𝟎𝟏𝟐𝟖

𝑷 𝒀≥𝟏
=𝟏−𝑷 𝒀=𝟎
𝟏 − 𝟎. 𝟗𝒏 > 𝟎. 𝟗𝟓
?
𝟎. 𝟗𝒏 < 𝟎. 𝟎𝟓
𝒏 > 𝟐𝟖. 𝟒
𝒏 ≥ 𝟐𝟗

You won’t be able to do (d) until

Chapter 2.
𝑛 or 𝑝 is unknown
S2 Jan 2013 Q3

𝑃 𝑋 = 0 = 1 − 𝑝 12 = 0.05
12
1 − 𝑝 = 0.05 = 0.779
𝑝 = 0.221 ?
Exercise 1C

Further Exercises (not in textbook)

1 I play a game 10 times with a 𝑝 probability of winning.
Given that the probability I win every time is 10%,
determine 𝑝.
𝒑𝟏𝟎 = 𝟎. 𝟏 ? 𝒑 = 𝟎. 𝟕𝟗𝟒
I play a different game 10 times with a 𝑞 probability of
2 winning. Given that 𝑃 𝑋 = 0 = 0.1, where 𝑋 is the
number of times I win, determine 𝑞
?
𝒒 = 𝟎. 𝟐𝟎𝟔
If a third game has a 40% chance of winning, how many
3
times will I have to play it to have at least a 99% chance of
winning more than once.
𝑷 𝑿 ≥ 𝟏 = 𝟏 − 𝟎. 𝟔𝒏 = 𝟎. 𝟗𝟗
?
𝒏 ≥ 𝟗. 𝟎𝟏𝟓 𝒏 = 𝟏𝟎
 Summary
We use a Binomial distribution when we want the probability of some number of
‘successes’ out of 𝑛 independent trials, in which each trial has only two outcomes.

These are all based on the parameters we set.

Description Name Params Outcomes Prob Func 𝑬𝑿 𝑽𝒂𝒓 𝑿
We count the number of Binomial Number of trials 𝑛 0, 1, … , 𝑛 𝑛 𝑥 𝑥 𝑛𝑝 𝑛𝑝 1 − 𝑝
𝑝 𝑥 = 𝑝 1−𝑝
‘successes’ after a Distribution 𝑥
number of trials, each 𝑩 𝒏, 𝒑 Probability of
with two outcomes
(‘success’ and ‘failure’). ?
success in each
trial 𝑝 ? ? ? ?
e.g. Number of heads
after 10 throws of an
unfair coin.

We can use tables to calculate 𝑃 𝑋 ≤ 𝑥 .

However, if 𝑝 > 0.5 we have to instead count the number of failures.