9.4 Expected Value and Applied Probabilistic Decision-Making

9.4 Expected Value and Applied Probabilistic Decision-Making

Probability theory reaches its most powerful practical application in Expected Value ($E(X)$) and probabilistic decision analysis. On the CLEP College Mathematics examination, expected value questions test your ability to model discrete probability distributions, calculate long-run average outcomes, evaluate whether games of chance are mathematically "fair," and analyze real-world commercial scenarios including insurance policy pricing, warranty cost-benefit choices, and business investment valuations.

---

1. Discrete Random Variables & Probability Distributions

A discrete random variable ($X$) is a variable that assumes a countable set of distinct numerical values resulting from a random phenomenon.

+-----------------------------------------------------------------------------+
|                   DISCRETE PROBABILITY DISTRIBUTION RULES                   |
|                                                                             |
|   A valid discrete probability distribution for random variable X assigns   |
|   a probability P(x_i) to each possible value x_i such that:                |
|                                                                             |
|   1. NON-NEGATIVITY:        0 <= P(x_i) <= 1  for every outcome x_i         |
|   2. TOTAL NORMALIZATION:   SUM [ P(x_i) ] = 1.0                            |
+-----------------------------------------------------------------------------+

Structure of a Probability Distribution Table

---

2. Expected Value Mathematical Mechanics

The Expected Value ($E(X)$ or $\mu$) of a discrete random variable is the probability-weighted average of all possible values that the variable can take.

The Expected Value Formula

$$E(X) = \mu = \sum_{i=1}^n [x_i \cdot P(x_i)] = x_1 P(x_1) + x_2 P(x_2) + \dots + x_n P(x_n)$$

Where:

• $x_i$ = Numerical payoff, gain, or loss associated with outcome $i$
• $P(x_i)$ = Probability of outcome $i$ occurring

Gross Expected Value vs. Net Expected Value

When participating in a game, raffle, or financial investment that requires an upfront entry cost or ticket price ($C$), you can calculate net expected value in two mathematically equivalent ways:

1. Method 1 (Net Payoffs in Distribution): Subtract the cost $C$ from each individual payoff $x_i$ before multiplying by probabilities: $$E(X_{\text{net}}) = \sum [ (x_i - C) \cdot P(x_i) ]$$
2. Method 2 (Gross Expected Payout Minus Cost): Compute the gross expected payout first, then subtract the entry cost: $$E(X_{\text{net}}) = E(X_{\text{gross}}) - C$$

---

3. Game Theory & The Concept of a "Fair Game"

In mathematical game theory and economics, the concept of fairness is defined strictly through expected value.

+-----------------------------------------------------------------------------+
|                        MATHEMATICAL FAIR GAME CRITERIA                      |
|                                                                             |
|   - FAIR GAME:           E(X_net) = 0   (Expected Payout == Cost of Entry)  |
|     Neither player nor house possesses a statistical long-term advantage.   |
|                                                                             |
|   - UNFAIR (HOUSE EDGE): E(X_net) < 0   (Player experiences expected loss)  |
|     Standard for casino games, lotteries, and charitable raffles.           |
|                                                                             |
|   - PLAYER ADVANTAGE:    E(X_net) > 0   (Player experiences expected gain)  |
+-----------------------------------------------------------------------------+

Casino House Edge: The American Roulette Wheel

An American roulette wheel contains $38$ equally sized pockets: numbers $1$ through $36$, plus $0$ and $00$. A player bets $\text{\textdollar}1$ on a single number. If the number hits, the casino pays $\text{\textdollar}35$ in profit (returning the $\text{\textdollar}1$ bet plus $\text{\textdollar}35$ winnings). If any other number hits, the player loses the $\text{\textdollar}1$ bet.

$$\begin{aligned} E(X_{\text{net}}) &= (+\$35) \times P(\text{Hit}) + (-\$1) \times P(\text{Miss}) \\ &= (35) \times \left(\frac{1}{38}\right) + (-1) \times \left(\frac{37}{38}\right) \\ &= \frac{35 - 37}{38} = -\frac{2}{38} = -\frac{1}{19} \approx -\$0.0526 \end{aligned}$$

• Interpretation: For every $\text{\textdollar}1.00$ wagered, the player expects to lose an average of $5.26$ cents. The casino's House Edge is exactly $5.26\%$.

Worked Example 1: Charity Raffle Valuation

Problem: A civic charity sells $2{,}000$ raffle tickets at $\text{\textdollar}10$ per ticket. The raffle awards the following prizes:

• One $1\text{st}$ prize: $\text{\textdollar}5{,}000$
• Two $2\text{nd}$ prizes: $\text{\textdollar}1{,}000$ each
• Five $3\text{rd}$ prizes: $\text{\textdollar}200$ each

What is the expected net monetary gain or loss for a person who purchases one ticket?

Solution (Using Method 2):

1. Calculate the total prize pool value: $$\text{Total Gross Prizes} = (1 \times 5{,}000) + (2 \times 1{,}000) + (5 \times 200) = 5{,}000 + 2{,}000 + 1{,}000 = \$8{,}000$$
2. Calculate the gross expected return per ticket: $$E(X_{\text{gross}}) = \frac{\text{Total Gross Prize Pool}}{\text{Total Tickets Sold}} = \frac{\$8{,}000}{2{,}000} = \$4.00$$
3. Subtract the ticket purchase price ($C = \text{\textdollar}10.00$): $$E(X_{\text{net}}) = E(X_{\text{gross}}) - C = \$4.00 - \$10.00 = -\$6.00$$

• Conclusion: Purchasing a ticket yields an expected net loss of $\text{\textdollar}6.00$ per ticket (representing an average charitable contribution of $\text{\textdollar}6.00$).

---

4. Actuarial Science & Commercial Insurance Underwriting

Insurance companies operate by aggregating independent risks across large populations. Actuaries use expected value to determine the pure risk premium necessary to cover claims, subsequently adding overhead loadings for administrative expenses, capital reserves, and target profits.

$$\text{Pure Actuarial Premium} = E(\text{Claims Payout}) = \sum_{i=1}^k [\text{Claim Amount}_i \times P(\text{Claim}_i)]$$ $$\text{Insurer Expected Profit} = \text{Charged Retail Premium} - \text{Pure Actuarial Premium}$$

Worked Example 2: Term Life Insurance Pricing

Problem: An insurance company offers a 1-year term life insurance policy with a death benefit of $\text{\textdollar}250{,}000$ to a $35$-year-old individual for an annual premium of $\text{\textdollar}450$. Actuarial mortality tables establish that the probability of a person in this demographic dying during the policy year is $0.0012$.

1. What is the insurance company's expected payout per issued policy?
2. What is the insurance company's expected profit per issued policy?

Solution:

1. Calculate expected claim payout: $$E(\text{Payout}) = \$250{,}000 \times 0.0012 = \$300.00$$
2. Calculate expected profit from the insurer's viewpoint: $$\text{Expected Profit} = \text{Premium Collected} - E(\text{Payout}) = \$450.00 - \$300.00 = \$150.00$$

• Interpretation: The company expects to earn an average profit of $\text{\textdollar}150$ on each policy issued to this demographic group.

---

5. Consumer Warranty & Commercial Risk Decisions

Consumers and corporate executives frequently use expected value to decide whether to purchase extended product warranties or undertake risky capital investments.

Worked Example 3: Extended Appliance Warranty Analysis

Problem: A consumer purchases a high-end refrigerator for $\text{\textdollar}2{,}000$. The retailer offers a 3-year extended warranty for $\text{\textdollar}250$. Industry reliability data indicates:

• A $10\%$ probability of major mechanical failure requiring complete $\text{\textdollar}2{,}000$ replacement.
• A $20\%$ probability of minor component failure requiring a $\text{\textdollar}300$ repair.
• A $70\%$ probability of zero defects over the 3-year term.

From a purely mathematical standpoint, is purchasing the $\text{\textdollar}250$ warranty beneficial to the consumer?

Solution:

1. Calculate the expected repair/replacement cost without a warranty ($E(\text{Loss})$): $$E(\text{Loss}) = (\$2{,}000 \times 0.10) + (\$300 \times 0.20) + (\$0 \times 0.70) = \$200 + \$60 + \$0 = \$260.00$$
2. Compare expected loss to the warranty cost:
   • Expected repair cost without warranty: $\text{\textdollar}260.00$
   • Cost of warranty: $\text{\textdollar}250.00$
   • Expected Net Benefit to Consumer $= \text{\textdollar}260.00 - \text{\textdollar}250.00 = +\text{\textdollar}10.00$

• Decision: Because the expected payout covered by the warranty ($\text{\textdollar}260$) exceeds the warranty price ($\text{\textdollar}250$), purchasing the warranty provides a positive expected net benefit of $\text{\textdollar}10.00$.

Capital Project Valuation: Expected Monetary Value (EMV)

In business decision-making under uncertainty, managers select projects maximizing Expected Monetary Value (EMV):

• Project Alpha EMV: $$E(\text{Alpha}) = (500{,}000 \times 0.30) + (200{,}000 \times 0.50) + (-100{,}000 \times 0.20) = 150{,}000 + 100{,}000 - 20{,}000 = \$230{,}000$$
• Project Beta EMV: $$E(\text{Beta}) = (300{,}000 \times 0.30) + (220{,}000 \times 0.50) + (50{,}000 \times 0.20) = 90{,}000 + 110{,}000 + 10{,}000 = \$210{,}000$$
• Decision: Project Alpha yields higher long-term expected value ($\text{\textdollar}230{,}000 > \text{\textdollar}210{,}000$), although Project Beta carries zero risk of negative cash flow in a recession.

---

6. TI-30XS MultiView Calculator Workflow

+-----------------------------------------------------------------------------+
|                   TI-30XS EXPECTED VALUE CALCULATION WORKFLOW               |
|                                                                             |
|   DIRECT ARITHMETIC STRING EVALUATION:                                      |
|   Input: 5000 * ( 1 / 2000 ) + 1000 * ( 2 / 2000 ) + 200 * ( 5 / 2000 )    |
|   Press [enter] -> displays 4                                               |
|   Then subtract ticket cost: ans - 10 [enter] -> displays -6                |
+-----------------------------------------------------------------------------+

---

7. Common CLEP Traps & Strategic Checkpoints

• Trap 1: Forgetting Negative Signs on Losses: When calculating net expected value, losses (such as paying an entry fee or suffering an investment loss) must be entered with a negative sign (e.g., $-\text{\textdollar}100 \times 0.20 = -\text{\textdollar}20$).
• Trap 2: Normalization Check Failure: Always verify that all probabilities in a distribution sum to exactly $1.0$. If the probabilities sum to less than or greater than $1.0$, the distribution is invalid or an outcome has been omitted.
• Trap 3: Confusing Expected Value with Most Likely Outcome: Expected value is a long-run mathematical average, which often does not equal any single possible outcome. For instance, the expected value of rolling a fair 6-sided die is $\frac{1+2+3+4+5+6}{6} = 3.5$, even though rolling a $3.5$ on a single roll is impossible.