Can Christians Do Probability?

• Probability talks about “chance”

• Does “chance” exist if God is in control of
the universe?
• Proverbs 16:33 The lot is cast into the lap,
but its every decision is from the LORD.
• Obviously, God doesn’t deal with “chance”
• What about us? Should we?
 An Example From Medicine *

• Nothing in the universe happens outside of

God’s power & control
• Let’s say someone can’t move their arm
• Why won’t it move? (multiple choice)
A. It is not God’s will that the arm should
move; it will move when God wills it.
B. A medical explanation involving muscles,
bones and joints.
* Thanks to Dr. David Beck
 What’s the Point
• Which answer is correct?
• Which should come from a doctor?
• Our goal: exactly the same!
• Provide a helpful explanation of how
random, unpredictable processes behave
• Chance is very useful concept when playing
Yahtzee or hearts
 Chapter 2: Probability
• Probability predicts the _________ of ______
occurrences (foundation of inferential statistics)
• Observation: any recording of information
– Not necessarily numerical
• Statistical Experiment: an activity or process whose
outcome is subject to _________ (e.g. survey, dice)
• Experimental, observational and retrospective
studies are all experiments
 Sample Space
• The sample space (denoted S) is the set of
____ possible _________ of an experiment
• Dice: S = { }
• Select a card: how many elements are in S?
• For multi-stage experiments, _____ diagrams
can help clarify the sample space
• Flip 3 coins, how many outcomes w/ 2 Heads?
 Tree Diagrams: Determining All
Outcomes of an Experiment
Quarter Nickel Dime Outcome

H
P(H) = 1/2

P(T) = 1/2
T
 When S Can’t be Enumerated
• Statement or rule method
• If S has too many outcomes to reasonably list:
• S = {x | x is a full-time residential student at LU}
• If S is continuous
• Time until nurse on 12 hr shift sees 1st patient
• S = { t | 0 ≤ t ≤ 12 }
• Sometimes S has no logical upper bound:
• Waiting time: S = { t | t ≥ 0}
• Number of customers:
 Events
• An event is any _________ of the sample space
• A collection of one or more outcomes
• Example: drawing a club
• Complement of event A, denoted A’, is the set of
all elements is S that are _______ in A

A A’
• Complement Rule: P(A) + P(A’) = ___
• Let A = Drawing a heart
• P(A) = P(A’)
 Complementary Events
• P(A) + P(A’) =
• Why this helps….
• Medic on call: P(0 calls) = 0.30
• P(1 call) = 0.35
• P(At least 2 calls) =
 Intersection of A and B (A ∩ B)
• The event containing all elements that are
________ to BOTH A and B

A B
• A and B are _________ __________ if they
have no elements in common: A ∩ B =
• Can’t both occur
A B C
 Union of A and B (i.e. A B)
• The event containing all elements that
belong to ________ A ___ B (or BOTH)

A B
• ex. A = {2,5,8} B = {2,3,4,5}
• What is A  B?
• What is A ∩ B?
Gets More Interesting with 3 Events
• (A  C’) ∩ B =
• Does (A ∩ B)  C = A ∩ (B  C)?

7 6
2
A 1
B
3
4
C
8
U 5
 More Complex Problems
• S = { a,b,c,d,e,f,g,h,i }
• A = {a,e,i} B = {b, d, e, h} and C ={c,d,e,f}
• Find A ∩ C
• Find (A  B)’
• Find A  (B ∩ C)’

• Find (A ∩ C)  (A ∩ B’)
 2.2 Axioms, Interpretations and
Properties of Probability
• We’ll be consider only f______ state spaces
• Each element in S is assigned a w_______ (or
probability) specifies its l_________ of occurrence
• The probability of event A is the _____ of the prob’s
of ______ sample points in A
• 0 ≤ P(A) ≤
• P( ) =
• P(S) =
• If A1, A2., …. are ___________ __________, then
P(A1 A2 …) = P(A1)+ P(A2)+ …..
 Examples and Assumptions
• P(selecting a face card) =
• Assumes that each outcome is ________ ________
– Text call the indifference approach
• Rule 2.3 If there are N ______ ________ outcomes
and n correspond to event A, P(A) =
• Ex. Roll 2 dice P(sum ≥ 10) =
• Poker: P(two pairs and one other card) = ??
– Start with P(2 aces and 2 kings) =
 Interpreting Probability
• How do we assign these probabilties?
• The objective interpretation (empirical) relies on
data and __________ frequencies:
– Examples:
• Common consensus belief: limiting relative
frequencies come from assumption that each
outcome is _______ _________
– Examples:
• Many probabilities are inherently ______
 Additive Rules
• Palindrome Hotel: “40% of our guests come to swim, 40% to
play golf and 40% to do nothing!”
• Is this possible? What is P(Swim or Golf)?

Guest Golf Swim Nothing

Anna
Otto
Hannah
Eve
Bob
 Formula for P(A B)
• Anna swims & golfs, but shouldn’t _____
_____ _______!
• Only 3 of 5 either swam or golfed
• P(S or G) =

General Rule of Addition:

P(A B) = P(A) + P(B) − P(A∩B)
 Examples & Corollaries
• P(Heart Face card) =
• What about P(Heart Diamond)?
• A & B ______ _______: P(A B) =
• If A1, A2., …, An are _______ _______, then
P(A1 A2 … An) = P(A1)+ P(A2)+ …..+P(An)
• If there are N outcomes, assumed to be equally
likely, then the probability of each is _______
 What about 3 Events?
• P(A  B  C) = P(A) + P(B) + P(C) + …

7 6
2
A 1
B
3
4
C
8
U 5
 2.3 Counting Techniques
• Car is available in 3 different styles and 5
different colors.
• How many selections are available?
• Could approach this w/ a ____ _________
• Product Rule for ordered pairs: 1st element
can be selected n1 ways & 2nd in n2 ways,
then the number of pairs is
 General Product Rule
• Given an ordered collection of k elements (a
k-tuple) and n1,…,nk outcomes of elements 1
through k, the number of outcomes for the
series of elements = n1 × n2 × … × nk
• How this works out depends on context of
the problem
• How one event effects subsequent events
 Tree Diagrams: 2 Balls from 2 Jars
(Independent) Outcome
Ball 2
Ball 1 11
1
1 2 12
3 13
21
1
2 2 22
3 23
• Get the same
diagram if we 1 31
use 1 jar w/ 3
2 32
replacement
3 33
2 Balls from 1 Jar w/ NO
replacement (Dependent)
Ball 2 Outcome
Ball 1
2 12
1
3 13

1 21
2
3 23

1 31
3
2 32
Multiplications and Permutations
• Problem 1: choose 1 of 10 balls from 4
different buckets (or from 1 bucket where
balls are replaced so picks are independent)
• # of outcomes =
• Problem 2: choose 4 of 10 balls from 1
bucket w/ NO replacement
• # of outcomes =
• This is called a permutation problem
 More Examples
• Problem 1: select a radio station name
– First letter must be K or W, other three letters
can be anything and letters can repeat
• Problem 2: Same as problem 1 except letters
aren’t allowed to repeat
 Permutation Formula
• nPn = n! = the # of ways to arrange n objects
• n! = n×(n−1) ×(n − 2) ×…×1 (and 0! = 1)
• Sometimes we might only be concerned
about the arrangement of r items
• Ex. How many 4-color codes can be made
from 7 different colors (no repeats)
• Formula: nPr =
 Extending Permutations
• How many ways can we arrange the letters in
the word redeye
• Easy case: each “e” is different (rEdeyε)
• 6! Arrangements
• What if the e’s are the same?
• Eεerdy, Eeεrdy, εEerdy, εeErdy, eEεrdy, and
eεErdy are now the same!
 More Complex Formula
• Distinct permutations n objects of which n1
are one kind, … nk are a kth kind is
𝑛!
• Formula:
𝑛1 !𝑛2 !…𝑛𝑘 !
• Arrangement of letters in banana
 Surprisingly Identical Problem
• How many ways can we partition n objects
into r subsets called cells
• Ex. How can we put 11 people into 4 rooms:
1 solo, 1 double, 2 rooms for four?
• Like arranging letters in mississippi
• Positions for each letter correspond to people
assigned to rooms!
 Special Case: Combinations
• How many ways can we partition n objects
into one cell with r and another with (n−r)?
𝑛!
• Formula:
𝑟! 𝑛−𝑟 !
• Denoted nCr: how many ways can we select r
out of n to be in one group?
 Which to Use? Does order (or
placement) matter?
• 8 Runners, Top 3 get medals → order
matters! use permutations:
8! 8  7  6  5  ... 1
n Pr    8  7  6  336
(8  3)! 5  4  ... 1
• 8 Runners, Top 3 advance to finals → order
doesn’t matter! use combinations:
8! 8 7  6
n Cr    8  7  56
3!(8  3)! 3  2 1
 From Permutations to
Combinations
• Permutations: simple application of
Fundamental Counting Principle (FCP) with
n1 = n 2 = and n3 =
• Combinations ignores order, so we _______
by the # of ways to arrange r objects.
• This is also an application of the FCP:
 Combinations vs. Permutations
• How to choose 4 out of 10 faculty to teach 4
different summer courses?
• How to choose 3 out of 12 students to be on a
committee?
• How to choose 5 out of 10 people to bring various
items to a class picnic?
• How to choose 3 out of 18 people to bring ice?
• How to choose 4 out of 20 people for various club
offices?
 2.4 Conditional Probability

• 50 freshman
• 10 are in Marching Band
• 6 are in Orchestra
• 4 are in both Band & Orchestra
• What’s P(Fr is in Orchestra)?
• What’s P(Fr is in Orchestra | Fr is in Band)?
 Conditional Probability Formulas
and Intuition
• P(Orch | Bnd) = P(Bnd ∩ Orch) / P(Bnd) =
• This makes sense:
• Denominator is the # of freshmen in the
“given” category
• Numerator is the # in both categories
• Sometimes, we don’t have counts:
• 80% get coffee; 70%: get donut & coffee, find
P(D|C) =
 Displaying Conditional
Probabilities: Contingency Tables
Percentage of Evangelicals
Continent Small Moderate Large Total
(under 5%) (5.0 –14.9%) (over 15%)
Asia 19 2 2 23
S. Amer 1 4 0 5
Africa 3 3 7 13
West 7 5 1 13
Total 30 14 10 54
P(Nation has large % Ev | African) =
P(Nat’n is Asian | Low % of Ev) =
 Independent Events

• Fair dice: P(rolling a 3) =

• P(rolling a 3 | previous roll was a 3) =
• Why? Each roll is __________
• We don’t care about previous results
• Implication: P(A|B) = P(A)
• A and B are independent if P(A|B) = P(A)
and P(B|A) = P(B)
Multiplication Rule: re-arranging
conditional probability
• If P(B|A) = P(A ∩ B)/P(A) then…
• P(A ∩ B) =
• Think sequentially: 1st A must happen, then
need likelihood B happens given A has occurred
• Ex. box of 12 chocolates and 3 have nuts
• P(Nut on 1st choc AND Nut on 2nd choc)
– Obviously no replacement
 Impact of Independence

• Def: P(A|B) = P(A) and P(B|A) = P(B)

• e.g. Lions make the playoffs if the Vikings
lose and the Bears lose
• P(Vikings lose) = 0.4, P(Bears lose) = 0.7
• Product rule: P(V ∩ B) = P(V) P(B|V)
• Independence: P(V ∩ B) =
 Extension to Multiple Events

• P(A1 ∩ A2 ∩ … ∩ Ak) = P(A1)

P(A2|A1)P(A3|A2∩A1)…P(Ak:Ak−1∩… ∩A1)
• Ex. Sampling medical board: 20 members,
including 7 dentists
• Find P(none of 3 selected will be dentists)
 Conditional Probability & the
Birthday Problem
• What’s the probability that we all have
different birthdays?
• What’s P(A & B have different b’days)?
P(A)P(B|A) =
• What’s P(A, B & C have different b’days)?
P(A)P(B|A)P(C|A and B) =
1 2 3 4 5 6 7 8 9 10 11 12
365 364 363 362 361 360 359 358 357 356 355 354
1.00 1.00 0.99 0.98 0.97 0.96 0.94 0.93 0.91 0.88 0.86 0.83
 Law of Total Probability
• 30% of blue collar voters support Obama, while
75% of white collar voters do
• Ohio: 55% blue collar
• What’s P(support Obama)?
• Answer by breaking it down into 2 groups
• P(Ob) = P(BC)×P(Ob|BC)+P(WC)×P(Ob|WC)
 Generalizing Total Probability
• Events A1, … , Ak must partition S (they are
mutually exclusive and _________)
– Nothing in S falls outside of A1,…Ak
• If so P(B) = σ𝑛𝑖=1 𝑃 𝐵 𝐴𝑖 𝑃(𝐴𝑖 )
• Ex. 12 chocolates, 3 w/ nuts
• You pick 2nd. What’s P(Nut on 2 choice)?
 Bayes’ Theorem: Setup
• Events A1, A2, …, Ak partition S
• We’re given P(B|Ai)
– Much like law of total probability
– Easy enough to find P(A)
• What if we want to find P(Ai|B)?
• Obama ex: Given P(Ob|BC) & P(Ob/WC)
• What if we want to find P(BC|Ob)?
 Bayes’ Theorem
• We know P(Ai|B) =
• Total Probability P(B) =
• So P(Ai|B) =
• Ex. Find P(Blue-collar|Obama)
 Common Application of Bayes’
Theorem
• Cheap medical test: always detects a disease
but also generates false positives
• Let P(positive | no disease) = 0.14
• Only 3% of the population has the disease
• Find P(disease | positive)