EC 285 - 2026-09-24 - Lecture 05

EC 285 - Lecture 05 - 2026-09-24

Introductory Statistics · Zara Liaqat · SB 106

Week 3 · 16:00-17:15 · 75 min · Full transcript · Flashcards

Overview

This lecture finished Chapter 3 (Probability), picking up after the counting rules from last time. It covered the complement and addition rules, then showed how to organise two events in a probability table or a tree diagram. From the table it built conditional probability, the multiplication rule and the test for statistical independence, all practised on a deck of cards and a used-car lot. It ended with the names for the cells of the table (joint and marginal probabilities) and with odds.

Picking up from last lecture

Last lecture's classical probability often needs the counting formulas to find the number of outcomes.

Counting formulas on the sheet

Permutations count ordered arrangements of x objects picked from n. Combinations count selections where order does not matter.

Pxn=n!(n−x)!Cxn=n!x!(n−x)!

where n is the number of objects available and x is the number picked.
On the formula sheet for: Midterm 1, Midterm 2 (not on the final's sheet).

The purple-car problem from last time is worked in full under #Complement rule.

Exam

"Both of these show up on your formula sheet, so you don't have to memorize the formula, but you should still know how to read this. If you don't know what C or P or n or x is, then the formula sheet is of no use."

Probability rules

Complement rule

The probability that A does not happen is 1 minus the probability that it does. You already used this last lecture on the purple-car problem.

Why: A and its complement are mutually exclusive (no overlap) and collectively exhaustive (together they are the whole sample space), so their probabilities must add to exactly 1.

P(A¯)=1−P(A)i.e.P(A)+P(A¯)=1

where A is any event and A¯ is its complement, everything in the sample space that is not in A.
Not on any formula sheet, so memorise it.

Example: At least one purple car

A dealer has 25 cars (10 red, 7 blue, 6 white, 2 purple), and the agent receives 4 of them at random. Find the probability that at least one is purple. In class the answer was about 30%.

  1. P(at least one purple)=1−P(no purple) -- complement rule, since "at least one" is awkward to count directly
  2. C425=25⋅24⋅23⋅224⋅3⋅2⋅1=30360024=12650 -- all possible sets of 4 cars
  3. C423=23⋅22⋅21⋅2024=21252024=8855 -- sets of 4 picked only from the 23 non-purple cars
  4. P(no purple)=885512650=0.7 -- classical probability
  5. 1−0.7=0.3 -- apply the complement rule

Answer: 0.3, or 30% -- there is a 30% chance the agent gets at least one purple car.

Addition rule

The probability of A or B (the union) is the probability of A plus the probability of B, minus the overlap. If A and B are mutually exclusive, the overlap is zero and you just add.

Why: Adding P(A) and P(B) counts the overlap area twice, so you subtract it once.

P(A∪B)=P(A)+P(B)−P(A∩B)

where A∪B means "A or B" and A∩B means "both A and B".
Not on any formula sheet, so memorise it.

For mutually exclusive events, it becomes:

P(A∪B)=P(A)+P(B)

where A and B have no outcomes in common.

A student asked whether "A or B" means it can't be both. That is only true when the events are mutually exclusive. With cards, it is not true, because some cards are both red and an ace.

Example: Red or ace

Pick one card from a standard 52-card deck. A is "the card is an ace" and B is "the card is red". Find P(A∪B).

  1. P(A)=452 -- there are 4 aces
  2. P(B)=2652 -- half the deck is red
  3. P(A∩B)=252 -- the two red aces
  4. P(A∪B)=452+2652−252 -- addition rule, taking the red aces out once so they aren't counted twice
  5. =2852=713≈0.538 -- simplify

Answer: 2852=713 -- a bit over half the deck is either red or an ace. Leaving it as 28/52 was fine in class.

Exam

"It is called the addition rule. It's good to know the term, the name of the formula." If a question asks for the probability that a card "is either red or ace", "it means I'm asking you to apply the addition rule."

Probability tables and tree diagrams

These are two ways to lay out two events so the rules above are easy to apply.

Probability table

A 2 by 2 grid for two events A and B. The four middle cells are the intersections. The last row and last column are the totals of each row or column, and the corner is 1.

Why: Each row adds up to its end value, and everything adds up to 1, so a few known numbers are enough to fill in the rest. Drawing it stops you making mistakes.

B B¯
A P(A∩B) P(A∩B¯) P(A)
A¯ P(A¯∩B) P(A¯∩B¯) P(A¯)
P(B) P(B¯) P(S)=1.0

For example, P(A)=P(A∩B)+P(A∩B¯). The four middle cells always add to 1 because they are mutually exclusive and they are the only possible outcomes.

Example: Table for a deck of cards

A standard deck has 52 cards: half red, half black, and 4 aces. A is "ace" and B is "red". Fill in the table with counts first, then turn them into probabilities.

  1. Corner =52 -- total cards
  2. Red =26, black =26 -- half and half
  3. Red aces =2, black aces =2, so aces =4 -- one ace per suit
  4. Red non-aces =26−2=24, black non-aces =26−2=24 -- the rest of each column
  5. Non-aces =24+24=48, which also equals 52−4 -- check the row total
  6. Divide every cell by 52, for example P(A∩B)=252 and P(A)=452 -- turn counts into probabilities

Answer: Middle cells are 252,252,2452,2452, with totals 452 (ace), 4852 (non-ace), 2652 (red), 2652 (black) -- this table is used for the rest of the card questions.

Example: Table for the used-car lot

Of the cars on a used car lot, 70% have air conditioning (AC), 40% have a CD player (CD), and 20% have both. Fill in the table.

  1. P(AC)=0.7, so P(AC―)=1−0.7=0.3 -- complement rule on the AC total
  2. P(CD)=0.4, so P(CD―)=1−0.4=0.6 -- complement rule on the CD total
  3. P(AC∩CD)=0.2 -- given
  4. P(AC∩CD―)=0.7−0.2=0.5 -- the AC row must add to 0.7
  5. P(AC―∩CD)=0.4−0.2=0.2 -- the CD column must add to 0.4
  6. P(AC―∩CD―)=0.3−0.2=0.1 -- the no-AC row must add to 0.3; check 0.5+0.1=0.6

Answer: Middle cells are 0.2, 0.5, 0.2 and 0.1 -- three numbers from the question were enough to fill the whole table.

It doesn't matter whether AC goes on the rows or the columns. The table looks different, but the answers are the same.

Exam

"I'm only going to test you on a 2x2 or a 3x3, but it should be just two events, A and B." She recommends a table every time: "For a question like this, it becomes very easy. If it's a multiple-choice question, you don't have to draw it properly. You just draw a quick sketch." And: "If I ask you to draw a table, then you'd have to." You should also know a deck has 52 cards, 4 suits, half red and half black, and 4 aces "for questions on the exam."

Tree diagram

The same information as a probability table, drawn as branches. The first split is one event (AC or no AC). Each branch then splits on the second event (CD or no CD), giving four end points, which are the four intersections.

Why: It shows the conditional probabilities along the branches, for example the AC to CD branch is 0.20.7 and the no-AC to CD branch is 0.20.3. The lecturer prefers the table as simpler.

Conditional probability and the multiplication rule

Conditional probability

The probability of A given that you already know B has happened, written P(A∣B) and read "A given B". The extra information usually changes the probability. For example, the chance a student passes, given that they studied.

Why: Once you know B happened, B becomes your new sample space. So you divide by P(B) instead of by the full sample space. The "also" part, A happening too, is the intersection.

P(A∣B)=P(A∩B)P(B)

where A∩B is both happening and B is the event you know has occurred.
Not on any formula sheet, so memorise it.

The condition, the event after the bar, always goes in the denominator. So P(B∣A)=P(A∩B)P(A), since A∩B is the same as B∩A.

Example: CD player given AC

Using the car table above, find the probability that a car has a CD player, given that it has AC.

  1. P(CD∣AC)=P(AC∩CD)P(AC) -- conditional probability; AC is the condition, so it goes on the bottom
  2. =0.20.7 -- read both numbers off the table
  3. =27≈0.286 -- simplify

Answer: 27, about 28.6% -- among cars that have AC, a bit under 3 in 10 also have a CD player.

Exam

"You should remember this formula. There will be questions too on the exam that asks you to compute the conditional probabilities." And: "You have to memorize this formula. If you understand the logic, it's easy." On notation, she said the bracket style doesn't matter: "I won't deduct any points whatever the brackets."

Multiplication rule

The probability of both A and B equals the probability of A given B, times the probability of B.

Why: It is just the conditional probability formula rearranged, so it is the same thing. The lecturer said she rarely remembers the multiplication rule itself and derives it from the conditional formula.

P(A∩B)=P(A∣B)P(B)

where P(A∣B) is the conditional probability and P(B) is the probability of the condition.
Not on any formula sheet, so memorise it (or derive it from conditional probability).

Made-up example: Studying and passing

60% of students studied, and 90% of the students who studied passed. Find the probability that a student both studied and passed.

  1. P(pass∩studied)=P(pass∣studied)P(studied) -- multiplication rule
  2. =0.9×0.6 -- put in the numbers
  3. =0.54 -- multiply

Answer: 0.54 -- 54% of all students both studied and passed.

Statistical independence

Checking for independence

Two events are statistically independent if and only if the probability of both equals the product of their individual probabilities. You get the individual probabilities from the table's totals and the intersection from the first middle cell.

Why: Independence means knowing B happened tells you nothing new about A, so P(A∣B)=P(A). Putting that into the multiplication rule gives the test below.

P(A∩B)=P(A)P(B)

where A and B are the two events. If the two sides match, they are independent; if not, they are not.
Not on any formula sheet, so memorise it.

Example: Ace and red

Using the card table, are "ace" (A) and "red" (B) independent?

  1. P(A∩B)=252 -- the red-ace cell
  2. P(A)P(B)=452×2652 -- multiply the two totals
  3. =452×12=252 -- since 26 out of 52 is one half
  4. 252=252 -- the two sides match

Answer: Yes, they are independent -- knowing a card is red doesn't change the chance it is an ace.

Example: AC and CD player

Using the car table, are AC and CD independent?

  1. P(AC∩CD)=0.2 -- from the table
  2. P(AC)P(CD)=0.7×0.4=0.28 -- multiply the two totals
  3. 0.2≠0.28 -- the two sides differ

Answer: No, they are not independent -- having AC changes the chance a car has a CD player.

Joint and marginal probabilities

These are new names for the parts of the probability table you already used.

Joint probability

The four middle cells of a 2 by 2 table. Each is the probability of two events happening together, an intersection such as AC and CD, or no AC and no CD.

Why: "Joint" means the two outcomes happen jointly.

P(A∩B)=number of outcomes satisfying A and Btotal number of elementary outcomes

where the top counts outcomes in both events and the bottom counts every outcome.
Not on any formula sheet, so memorise it.

Example: A red ace

Find the joint probability that a card is both an ace and red.

  1. Outcomes that are both ace and red =2 -- ace of hearts and ace of diamonds
  2. P(Ace∩Red)=252=126 -- divide by all 52 cards

Answer: 252 -- the red-ace cell of the card table.

Marginal probability

The totals at the ends of the table (the margins). You get one by adding the joint probabilities across its row or down its column.

Why: The events you add over split the whole sample space with no overlap, so adding the pieces gives the whole probability of A.

P(A)=P(A∩B1)+P(A∩B2)+⋯+P(A∩Bk)

where B1,…,Bk are mutually exclusive and collectively exhaustive events.
Not on any formula sheet, so memorise it.

Example: Probability of an ace

Find the marginal probability that a card is an ace, using red and black as the split.

  1. P(Ace)=P(Ace∩Red)+P(Ace∩Black) -- red and black cover the whole deck with no overlap
  2. =252+252 -- read off the ace row
  3. =452 -- add

Answer: 452 -- the total at the end of the ace row.

Exam

"Remember the terms. I may use them. Joint refers to the intersections, the four cells in the middle for a 2x2 diagram. And then the ends, where we have the totals, are marginal probabilities."

Odds

Odds ratio

The odds in favour of an event are its probability divided by the probability of its complement: success over failure. Odds of 3 to 1 mean you win three for every one you lose.

Why: Odds compare the chance it happens against the chance it doesn't, instead of against everything. That's the idea behind "May the odds be ever in your favor."

odds=P(A)1−P(A)=P(A)P(A¯)

where P(A) is the probability of the event and P(A¯) is the probability it doesn't happen.
Not on any formula sheet, so memorise it.

Example: Probability from odds of 3 to 1

The odds of winning are 3 to 1. Find the probability of winning, PW.

  1. 3=PW1−PW -- put the odds into the formula
  2. 3(1−PW)=PW -- multiply both sides by 1−PW
  3. 3−3PW=PW -- expand
  4. 3=4PW -- add 3PW to both sides
  5. PW=34=0.75 -- divide both sides by 4
  6. Check: 0.750.25=3 -- gives back odds of 3 to 1

Answer: 0.75, or 75% -- not 25%, 33% or 50%, which were common wrong guesses in class.

Exam

"Odds are measured as a number, not a percent." A probability must come out below 1. She said the question can also go the other way: "I could give you probability of winning, and then ask you to find odds ratio."

Exam checklist

Admin & deadlines


Transcribed automatically from the lecture recording. Slides used: EC 285 - Ch. 03 - Probability Topics (slides).pdf. Course info used: EC 285 - Formula Sheet - Final.pdf, EC 285 - Formula Sheet - Midterm 1.pdf, EC 285 - Formula Sheet - Midterm 2.pdf, EC 285 - Syllabus - Fall 2026.pdf. Audio archived at /mnt/porsche/configs/lectures/archive/2026-09/EC 285 - 2026-09-24 - Lecture 05.opus.