Math 32 - 07 - Binomial Distribution

You might also like

Download as pdf or txt
Download as pdf or txt
You are on page 1of 7

Math 32 Lecture 6: Binomial Distribution Spring 2021

1 Factorial

Due to the frequent use in advanced mathematics, we define the factorial as

n
Y
n! = i
i=1
1! = 1

2! = 2 × 1 = 2

3! = 3 × 2 × 1 = 6

4! = 4 × 3 × 2 × 1 = 24

5! = 5 × 4 × 3 × 2 × 1 = 120

... = ... = ...

Also, mathematicians define 0! = 1 for later calculations.

1
Math 32 Lecture 6: Binomial Distribution Spring 2021

2 Bernoulli Trials

To continue our exploration of discrete distributions, we will look at situations


that have two disjoint possibilities.

TRUE FALSE

working broken

success failure

survived died
For math symbols to represent a Bernoulli trial, the events {1, 0} have
respective probabilities p and 1 − p.

For example, for one flip of a coin


P (heads) = p, P (tails) = 1 − p

2
Math 32 Lecture 6: Binomial Distribution Spring 2021

3 Arrangements

• Permutations (and the number of permutations) are the arrangements


when order matters
• Combinations (and the number of combinations) are the arrangements
when order does not matter

Example 1: Flipping 3 fair coins, what is the probability that heads will be
observed exactly twice?

{HHH, HHT, HT H, T HH,HT T, T HT, T T H, T T T }

Observe that there are 4 combinations, and


• order matters (permutations) for visualizing the sample space
• selections made with replacement
• 1 permutation with 3 heads (probability: 1 × [P (H)]3 [P (T )]0 )
• 3 permutations with 2 heads (probability: 3 × [P (H)]2 [P (T )]1 )
• 3 permutations with 1 head (probability: 3 × [P (H)]1 [P (T )]2 )
• 1 permutation with zero heads (probability: 1 × [P (H)]0 [P (T )]3 )

Here, the answer is 3 out of 8 possibilities.

3
Math 32 Lecture 6: Binomial Distribution Spring 2021

4 Choose

 
n n!
=
k k!(n − k)!

• said “n choose k”
• This choose operator keeps track of the number
of arrangements
• note 0! = 1

4
Math 32 Lecture 6: Binomial Distribution Spring 2021

5 Binomial Distribution

We say that the probability of observing exactly k instances, where each


instance had a prior probability of p in a binomial setting, is

 
n k
P (x = k) = p (1 − p)n−k
k
•0 ≤ k ≤ n
•0 ≤ p ≤ 1

Example 2: Historically, Squirtle defeats Charizard 32% of the time. If there


are 5 battles, what is the probability that Squirtle wins exactly 2 times?

Chat query: what are the values of n, k and p?


•n=5
•k=2
• p = 0.32
Thus the requested probability is

 
5 5!
P (k = 2) = (0.32)2 (1 − 0.32)5−2 = (0.32)2 (0.68)3 ≈ 0.3220
2 2!(5 − 2)!

5
Math 32 Lecture 6: Binomial Distribution Spring 2021

Example 3: Historically, Charizard defeats Squirtle 68% of the time. If


there are 5 battles, what is the probability that Charizard wins exactly 3
times?

•n=5
•k=3
• p = 0.68
• probability that Charizard wins 3 times: (0.68)3
• probability that Squirtle wins the other two battles: (0.32)2
• number of ways those 3 Charizard wins could happen: 53 = 10


Thus the requested probability is

 
5
P (k = 3) = (0.68)3(1 − 0.68)2 ≈ 0.3220
3

6
Math 32 Lecture 6: Binomial Distribution Spring 2021

6 Symmetry

The previous two examples had the same answer, which is true due to a
symmetry property in the choose operator:

   
n n
=
k n−k

is the same probability as

Q: How do we pick between p and 1 − p?


At first, it does not matter how you define the binomial setting for what
corresponds to p and what corresponds to 1−p, but you need to be consistent
in the rest of the task for how you defined your variables and use the value(s)
for k.

You might also like