BUS 660 Topic 1 payoff table analysis example

Quantitative Methods Grand Canyon University Free custom sample in 24 to 48h

This BUS 660 Topic 1 sample paper builds a payoff table for a composite regional bakery choosing between a large expansion, a small expansion and doing nothing. It was written for GCU BUS 660, Quantitative Methods, catalog code BUS-660, a core course in Grand Canyon University's MBA. With demand probabilities of 0.3, 0.5 and 0.2 drawn from ten years of regional data, expected value favors the large expansion at 149 thousand dollars, maximin favors doing nothing and minimax regret favors the small expansion. Perfect information would be worth 64 thousand. Because a 260-thousand loss would breach its covenant, the bakery chooses the small expansion.

CourseBUS 660 Quantitative Methods
TopicTopic 1
Paper typePayoff table analysis
LengthAbout 1,000 words, 2 pages
FormatAPA 7 student paper
SchoolGrand Canyon University (GCU)
ProgramMBA
UpdatedOctober 2026

Free sample paper for BUS 660 Topic 1

1

Should a Firm That a Bad Year Would Ruin Maximize the Average? A Payoff Table for a Bakery's Expansion Decision

[Author Name]

Colangelo College of Business, Grand Canyon University

BUS-660: Quantitative Methods

Topic 1 Payoff Table Analysis

[Instructor Name]

[Date]

The organization, setting and figures below are a composite written as a model document. No real employer, client, colleague or patient is described.

What this page is doingThe title states the commercial question the analysis ends on. The bakery and every figure are a composite; the sources are real.
2

Should a Firm That a Bad Year Would Ruin Maximize the Average? A Payoff Table for a Bakery's Expansion Decision

The Decision

A composite regional commercial bakery, which supplies bread to grocery stores and restaurants, must decide how to respond to rising orders. It can build a large expansion, adding a second production line; a small expansion, adding an oven and extending shifts; or do nothing. What it cannot control is how demand develops over the next three years. This analysis builds a payoff table, applies several decision criteria, and then asks whether a firm that would be ruined by the worst outcome should choose the option with the highest average. All amounts are illustrative, stated in thousands of dollars of added profit across three years.

Alternatives and States of Nature

A payoff table separates two things that are often merged: the alternatives the decision maker controls and the states of nature it does not (Render et al., 2018). The alternatives are large expansion, small expansion and do nothing. The states are high, medium and low demand growth. Each cell holds the payoff of one alternative under one state, and every payoff traces to an assumption. A large expansion under high demand earns 420, because the new line runs near capacity; under low demand it loses 260, because the bakery carries the line's fixed costs and loan payments with little extra volume. A small expansion earns less in good conditions and loses less in bad. Doing nothing earns 60 under every state, from existing growth within current capacity.

Table 1

Payoff Table, Thousands of Dollars Over Three Years

AlternativeHigh demandMedium demandLow demandExpected value
Large expansion420150Minus 260149
Small expansion240120Minus 40124
Do nothing60606060
Probability0.30.50.2

Where the Probabilities Came From

The probabilities are not given; they are estimated, and the analysis says how. Over the past ten years, regional bread demand grew strongly in three, moderately in five and weakly in two, which suggests 0.3, 0.5 and 0.2. Two large grocery chains have signaled more orders, which supports the moderate-to-high outcomes, but a new competitor has announced a plant nearby, which keeps the low outcome plausible. The probabilities are judgments, and the analysis tests how sensitive the choice is to them.

Expected Value and Other Criteria

Expected value weights each payoff by its probability. For the large expansion, 0.3 times 420 plus 0.5 times 150 plus 0.2 times minus 260 gives 149. The small expansion gives 124 and doing nothing 60. By expected value, the large expansion is best.

Table 2

Choices Under Four Criteria, Thousands of Dollars

CriterionLargeSmallDo nothingChoice
Expected value14912460Large
Maximax42024060Large
MaximinMinus 260Minus 4060Do nothing
Maximum regret320180360Small

Other criteria give different answers. The optimistic maximax criterion chooses the alternative with the best possible outcome, the large expansion at 420. The pessimistic maximin criterion chooses the alternative whose worst outcome is best, doing nothing at 60. The minimax regret criterion compares each payoff with the best payoff available in that state; the large expansion's maximum regret is 320, in the low-demand state, while the small expansion's is 180, so minimax regret favors the small expansion.

The Value of Better Information

If the bakery knew which state would occur, it would choose the best alternative for each: large under high demand, large under medium and do nothing under low. The expected payoff with perfect information is 0.3 times 420 plus 0.5 times 150 plus 0.2 times 60, or 213. The expected value of perfect information is 213 minus 149, or 64 thousand dollars. That is the most it would be worth paying for a demand study, and a study that could only partly resolve the uncertainty would be worth less.

Sensitivity to the Probabilities

The expected value ranking depends heavily on the low-demand probability. If the new competitor's plant makes low demand more likely, say 0.3 instead of 0.2, with high and medium demand at 0.25 and 0.45, the large expansion's expected value falls to about 95 and the small expansion's to 102. The small expansion then wins even on the average. A ranking that reverses with a ten-point change in one judgment is a weak basis for committing to the riskier option, and that sensitivity reinforces the recommendation below.

Should the Bakery Maximize the Average?

Expected value suits a firm that makes many similar decisions and can absorb a bad outcome, because over many decisions the averages prevail. The bakery is not in that position. It has about 200 thousand dollars of cash reserves and a loan covenant requiring it to stay profitable. A 260-thousand-dollar loss under low demand would breach the covenant and could force a sale of the business. For a firm that a bad outcome would ruin, the average is not the relevant number; avoiding ruin is. Kahneman and Tversky (1979) showed that people weigh losses more heavily than equal gains, and while that is often treated as a bias, for a firm facing ruin the asymmetry is real, not psychological.

Where Minimax Regret Comes From

The regret criterion used above measures, for each state of nature, how much worse an alternative does than the best alternative for that state, and chooses the alternative whose largest regret is smallest. Savage (1951) proposed it as a way of deciding without probabilities that avoids the extreme pessimism of maximin. For the bakery, regret favors the small expansion because it never leaves the bakery far behind the best choice in any state, which is a reasonable property for a firm that cannot afford to be badly wrong.

Recommendation

The analysis recommends the small expansion. It gives up 25 thousand dollars of expected value compared with the large expansion but limits the worst outcome to a 40-thousand-dollar loss, which the bakery can survive, and it is the choice under minimax regret. If demand proves high over the next year, the bakery can revisit the large expansion with better information and a stronger balance sheet.

References

Kahneman, D., & Tversky, A. (1979). Prospect theory: An analysis of decision under risk. Econometrica, 47(2), 263-291. https://doi.org/10.2307/1914185

Render, B., Stair, R. M., Hanna, M. E., & Hale, T. S. (2018). Quantitative analysis for management (13th ed.). Pearson.

Savage, L. J. (1951). The theory of statistical decision. Journal of the American Statistical Association, 46(253), 55-67. https://doi.org/10.1080/01621459.1951.10500768

The BUS-660 Topic 1 assignment instructions

Topic 1 in BUS-660 asks MBA students to analyze a decision under uncertainty with a payoff table. Strong analyses separate the alternatives a decision maker controls from the states of nature it does not, trace each payoff to an assumption, and explain where the probabilities came from. Students should apply expected value and other criteria such as maximax, maximin and minimax regret, compute the value of perfect information, test sensitivity to the probabilities and ask whether a firm that a bad outcome would ruin should maximize the average. Students should also apply minimax regret and explain where it came from, since it often favors a different choice than expected value.

How the BUS-660 Topic 1 example is put together

The finished example separates two things students merge, the alternatives a decision maker controls and the states of nature they do not. The table is built with one row per choice and one column per state, and every cell carries a payoff the reader can trace to an assumption. Expected value is computed across the states using stated probabilities, and the example says where those probabilities came from rather than treating them as given. It then compares the expected value criterion against the pessimistic and optimistic alternatives, and asks the question that matters commercially: whether a firm that would be ruined by the worst outcome should maximize the average at all.

Reading the BUS 660 Topic 1 rubric

Faculty reward a payoff table that ends in a commercial judgment. An Excellent paper tables three alternatives against three demand states with stated probabilities. Credit goes to expected values of 149, 124 and 60 and to choices under four criteria. The rubric values an expected value of perfect information of 64 thousand dollars and a sensitivity test that reverses the ranking. Strong papers explain why a firm with 200 thousand dollars of reserves cannot absorb a 260-thousand loss, with Kahneman and Tversky on losses. Top marks go to recommending the small expansion. Papers that compute the expected value of perfect information and compare it with the cost of a real market study show the table used to price information, not only to rank options.

BUS 660 Topic 1 help before you submit

Computing expected value and stopping is the standard shortfall, since the topic is about how a decision should be made and the arithmetic is the easy part. A second failure is probabilities inserted with no source, which makes the whole calculation an elaborate expression of a guess. Papers lose marks for states of nature that overlap or leave gaps, because the expected value is only meaningful if the states are mutually exclusive and cover everything. Recommending expected value for a firm that would be bankrupted by one of the outcomes ignores that averages apply over repetition and a company gets one attempt. Payoff figures given without a derivation leave nothing to verify, so a reader has no way to rely on the recommendation built from them.

Get a BUS-660 Topic 1 example written to your instructions

Send the BUS-660 Topic 1 problems and the rubric posted in your classroom, with the decision scenario your section supplied. We write a custom example to those criteria, with alternatives separated from states, payoffs traced to assumptions, probabilities sourced and the decision criterion itself argued, in 24 to 48 hours. The first is free.

More BUS 660 papers and MBA samples

BUS 660 Topic 1 questions, answered

When is expected value the wrong criterion?

When the decision happens once and one of the outcomes would end the business. Expected value describes what you would average over many repetitions, and a firm facing a single bet cannot rely on the average arriving. Where one branch is ruinous, a criterion weighted toward avoiding the worst case is more defensible, and saying so is what the topic is examining.

Where do the probabilities come from?

Historical frequency where the situation has recurred, published data where an external event is involved, or informed judgment where neither exists. All three are legitimate and they carry different weight, so name which you used. What is not defensible is a probability that appears in the table with no origin, since every conclusion afterward is a function of it.

How do I know my states of nature are complete?

They must be mutually exclusive and exhaustive, meaning exactly one will occur and no possibility is missing. Test it by asking whether any plausible future falls outside your columns, and whether two of them could happen together. Both failures break the expected value calculation, and the second is easier to commit than students expect when the states are described loosely.

Where can I find a free BUS 660 Topic 1 sample paper?

The payoff table analysis is shown here complete: the decision, alternatives and states, where the probabilities came from, expected value and other criteria, the value of information, sensitivity, whether to maximize the average and the recommendation. Describe the decision in your case and we will build the table free.

Why not choose the highest expected value in BUS 660 Topic 1?

The large expansion's 260-thousand-dollar loss under low demand would breach the bakery's loan covenant and could force a sale. For a firm a bad outcome would ruin, avoiding ruin matters more than the average.