Assessing Uncertainty and Risk
What this domain tests
Candidates must reason under uncertainty using expected value, probability ranges, risk tolerance, and sensitivity analysis. Know that expected value is the probability-weighted average of outcomes and is not the same as the most likely outcome. Know the difference between risk (known probabilities) and uncertainty (unknown or contested probabilities).
Understand risk aversion (a certain amount preferred over a gamble of equal expected value), loss aversion, overconfidence in ranges (ranges set too narrow), and the planning fallacy (systematic underestimation of time and cost). Know that diversification reduces idiosyncratic risk but not market-wide risk.
Traps in questions
- Most-likely outcome mistaken for expected value: choosing the modal outcome when a low-probability large loss dominates.
- Narrow confidence ranges: treating a point estimate as certain.
- Ignoring tail risk: dismissing low-probability catastrophic outcomes.
- Risk-neutral assumption: applying expected value when the decision-maker cannot absorb the downside.
Worked example
Option A pays 100,000 for sure. Option B pays 300,000 with probability 0.4 and 0 with probability 0.6. Expected value of B is 120,000, above A. But a firm that would be insolvent after a zero outcome may rationally take A despite the lower expected value. The right answer depends on the decision-maker's capacity to absorb loss, not on expected value alone. Always state the assumption.
Rule of thumb
Compute expected value first, then run a sensitivity check: does the ranking flip if the probability moves by 10 points? If yes, the decision is fragile and needs more analysis or a hedge.