Expected Value (Mean): The weighted average of all possible values of a random variable, weighted by their probabilities.
Formula: E[X] = Σ [x × P(X = x)] for all possible values x
- List all possible values of X
- Find probability for each value
- Multiply each value by its probability
- Sum all the products
Rolling a die: {1, 2, 3, 4, 5, 6}
Each outcome has probability 1/6
1×(1/6) = 1/6, 2×(1/6) = 2/6, 3×(1/6) = 3/6
4×(1/6) = 4/6, 5×(1/6) = 5/6, 6×(1/6) = 6/6
E[X] = 1/6 + 2/6 + 3/6 + 4/6 + 5/6 + 6/6
= (1 + 2 + 3 + 4 + 5 + 6)/6 = 21/6 = 3.5
The expected value is 3.5, even though you cannot roll a 3.5 on a die.
This represents the long-run average if the die is rolled many times.
The expected value of rolling a fair six-sided die is 3.5.
This represents the theoretical average outcome over many rolls.
• Expected Value Formula: E[X] = Σ [x × P(X = x)]
• Linearity: Expected value is a linear operator
• Interpretation: Theoretical long-run average
Net Winnings: Gross winnings minus cost of participation.
Good Investment: If expected value is positive, it's favorable; if negative, it's unfavorable.
Win $500: Net = $500 - $2 = $498, P = 1/1000
Win $100: Net = $100 - $2 = $98, P = 1/500 = 2/1000
Win $10: Net = $10 - $2 = $8, P = 1/100 = 10/1000
No win: Net = $0 - $2 = -$2, P = 1 - (1+2+10)/1000 = 987/1000
498 × (1/1000) = 0.498
98 × (2/1000) = 0.196
8 × (10/1000) = 0.08
(-2) × (987/1000) = -1.974
E[X] = 0.498 + 0.196 + 0.08 + (-1.974)
= 0.774 - 1.974 = -1.2
Expected net loss of $1.20 per ticket
This is not a good investment from a financial perspective
The expected net winnings are -$1.20 per ticket.
Buying the ticket is not a good investment as you expect to lose money on average.
• Net Value Calculation: Gross winnings minus cost
• Expected Value Formula: E[X] = Σ [x × P(X = x)]
• Investment Decision: Positive EV = good, Negative EV = bad
Sample Space: All possible outcomes of an experiment.
Probability Assignment: Calculating probability for each outcome.
Sample space: {HH, HT, TH, TT} (4 total outcomes)
All outcomes equally likely with probability 1/4
Both heads (HH): Win $5, P = 1/4
Exactly one head (HT, TH): Win $2, P = 2/4 = 1/2
Both tails (TT): Lose $3, P = 1/4
5 × (1/4) = 5/4 = 1.25
2 × (1/2) = 1
(-3) × (1/4) = -3/4 = -0.75
E[X] = 5/4 + 1 + (-3/4) = 5/4 + 4/4 - 3/4 = 6/4 = 1.5
The expected value of playing this game is $1.50.
This is a favorable game to play as you expect to win money on average.
• Sample Space Analysis: Enumerate all possible outcomes
• Expected Value Formula: E[X] = Σ [x × P(X = x)]
• Grouping Outcomes: Combine outcomes with same value
Expected Value (Mean): The long-run average value of repetitions of the same experiment.
Weighted Average: Each value weighted by its probability of occurrence.
Population Parameter: Theoretical mean of the probability distribution.
Balance Point: Where the probability distribution would balance if made into a physical object.
• Expected Value: E[X] = Σ [x × P(X = x)]
• Linear Transformation: E[aX + b] = aE[X] + b
• Sum of RVs: E[X + Y] = E[X] + E[Y]
• Constant: E[c] = c
Total Income: Base salary + (Commission per sale × Number of sales).
Linear Transformation: If Y = aX + b, then E[Y] = aE[X] + b.
Income Y = $2000 + $500 × Number of sales (X)
So Y = 500X + 2000
E[X] = Σ [x × P(X = x)]
= 0×0.1 + 1×0.2 + 2×0.4 + 3×0.2 + 4×0.1
= 0 + 0.2 + 0.8 + 0.6 + 0.4 = 2.0
E[Y] = E[500X + 2000] = 500E[X] + 2000
= 500 × 2.0 + 2000 = 1000 + 2000 = $3000
E[Y] = Σ [y × P(Y = y)]
= 2000×0.1 + 2500×0.2 + 3000×0.4 + 3500×0.2 + 4000×0.1
= 200 + 500 + 1200 + 700 + 400 = $3000 ✓
The expected monthly income is $3000.
This includes the base salary of $2000 plus expected commission of $1000.
• Linear Transformation: E[aX + b] = aE[X] + b
• Expected Value Formula: E[X] = Σ [x × P(X = x)]
• Verification: Alternative calculation confirms result
Comparative Analysis: Calculate expected value for each option and compare.
Decision Making: Higher expected value indicates better option under uncertainty.
E[A] = $10,000 × 0.6 + $5,000 × 0.4
= $6,000 + $2,000 = $8,000
E[B] = $15,000 × 0.3 + $8,000 × 0.5 + $2,000 × 0.2
= $4,500 + $4,000 + $400 = $8,900
E[A] = $8,000, E[B] = $8,900
Since $8,900 > $8,000, Investment B has higher expected return
While B has higher expected return, also consider risk and variance.
Investment B has wider range of possible returns ($2,000 to $15,000).
Investment B has the higher expected return of $8,900 compared to Investment A's $8,000.
Based on expected value alone, Investment B is the better choice.
• Expected Value Comparison: Calculate for each option and compare
• Decision Making: Higher expected value indicates better option
• Risk Consideration: Expected value doesn't account for risk/volatility
Complete Summary: Expected Value (Mean) of Discrete Random Variables
Expected Value: The weighted average of all possible values, weighted by their probabilities.
Formula: E[X] = Σ [x × P(X = x)] for all possible values x.
Notation: Also denoted as μ (mu) or mean of the distribution.
Interpretation: Theoretical long-run average if experiment is repeated infinitely.
Linearity: E[aX + b] = aE[X] + b for constants a, b.
Constants: E[c] = c where c is any constant.
Sum Rule: E[X + Y] = E[X] + E[Y] for any two random variables.
Doesn't Require Possible Value: Expected value may not be an actual outcome.
Step 1: List all possible values of X.
Step 2: Find probability for each value P(X = x).
Step 3: Calculate each product x × P(X = x).
Step 4: Sum all products to get E[X].
Verify Probabilities: Ensure Σ P(X = x) = 1 before calculating E[X].
Use Linearity: For Y = aX + b, compute E[Y] = aE[X] + b.
Group Values: Combine outcomes with same value to simplify calculations.
Check Reasonableness: Verify result makes sense in context.
Expected Value (Mean): The theoretical average value of a random variable over many repetitions of an experiment. It's the center of the probability distribution and represents the long-run average if the experiment could be repeated infinitely.
Discrete Random Variable: A variable that can take on a countable number of distinct values, typically integers or specific real numbers.
Probability Mass Function (PMF): The function P(X = x) that gives the probability that the random variable X takes on the value x.
Weighted Average: An average where each value contributes proportionally to its probability of occurrence, rather than equally as in a simple average.
Long-Run Average: The concept that if an experiment is repeated many times, the average of the observed values will approach the expected value.
- Identify the Random Variable: What quantity are you measuring?
- List All Possible Values: What values can X take?
- Determine Probabilities: Find P(X = x) for each possible x.
- Verify Distribution: Confirm that probabilities sum to 1.
- Calculate Products: Compute x × P(X = x) for each x.
- Sum Products: E[X] = Σ [x × P(X = x)].
• Basic Formula: E[X] = Σ [x × P(X = x)]
• Linear Transformation: E[aX + b] = aE[X] + b
• Sum of Variables: E[X + Y] = E[X] + E[Y]
• Constant: E[c] = c
Questions & Answers
Question: Why is the expected value of a die roll 3.5 when you can never actually roll a 3.5?
Answer: This is a common source of confusion! The expected value is not a prediction for a single roll, but rather the theoretical average over many rolls.
Understanding the concept:
- If you roll a die 6 times, you might get {1, 2, 3, 4, 5, 6}, average = 3.5
- If you roll it 600 times, the average of all outcomes will approach 3.5
- It's the "balance point" of the probability distribution
The expected value represents what would happen on average if the experiment could be repeated infinitely many times. It's a measure of the center of the distribution, not a possible outcome.
Think of it like the center of gravity of the distribution - where it would balance if made into a physical object.
Question: In Exercise 5, why did you say Investment B is better just because it has a higher expected return? What about risk?
Answer: Excellent observation! You're absolutely right that expected value alone doesn't tell the whole story.
Limitations of Expected Value:
- Expected value only measures the center, not the spread or variability
- Investment B has wider range ($2,000 to $15,000) than A ($5,000 to $10,000)
- Higher variability means higher risk
Additional Considerations:
For complete analysis, you'd also want to consider variance/standard deviation, potential losses, and your risk tolerance. Expected value maximization is optimal for risk-neutral investors, but risk-averse individuals might prefer lower expected return with less variability.
However, among investments with similar risk levels, higher expected value is generally preferred.
Question: Why does E[aX + b] = aE[X] + b? Can you explain the math behind this property?
Answer: This is called the linearity property of expectation, and it's quite elegant:
Mathematical proof:
Let Y = aX + b, then:
E[Y] = Σ [y × P(Y = y)]
But since Y = aX + b, when X = x, then Y = ax + b
So: E[aX + b] = Σ [(ax + b) × P(X = x)]
= Σ [ax × P(X = x)] + Σ [b × P(X = x)]
= a × Σ [x × P(X = x)] + b × Σ P(X = x)
= a × E[X] + b × 1 (since probabilities sum to 1)
= aE[X] + b
This property is extremely useful because it allows us to easily find the expected value of transformed random variables without recalculating the entire distribution.