Analyst Dojo
Lesson 5 of 7 11–14% of the exam

Scenario analysis with probability weights

Re-run the whole forecast model under a few coherent scenarios, each giving its own equity value. Weight those values by your subjective probabilities, counting any negative value as zero, and compare the weighted value with market capitalization. Sensitivity analysis instead changes chosen inputs while holding everything else fixed.

5 min cram, 22 min full read. Based on the 2027 CFA® Program Level I curriculum, Equities, Learning Module 8. Sitting in November 2026? The 2026 version of this lesson is still available.

Tailor this to you: take the self-assessment

Before this: Asset and financing models

On the exam

  • Explain the rationale, construction, and uses of financial statement forecast models in equity valuation
  • Estimate and interpret the value of an equity security based on the outputs of a financial statement forecast model

Interpret a probability-weighted equity value. Expect: a weighted value from scenario equity values (with a negative scenario set to zero), whether a share looks undervalued, overvalued or fairly valued against market capitalization, what the probabilities are (subjective, not true), and scenario versus sensitivity analysis.

How deeply to learn each part
Know the idea
Understand it in general. It frames other topics, but you will not calculate with it.
Use it
Understand it well enough to solve a question with it. Recognize when it applies.
Memorize
Learn it by heart. The exam will not give it to you.

You know this one. The full lesson is folded away. The takeaways, formulas and checkpoint below are your quick refresher, and the checkpoint is a fair test of whether you still do.

You tested strong here. The full lesson is folded away. The takeaways, formulas and checkpoint below are your quick refresher, and the checkpoint is a fair test of whether you still know it.

Cram mode: the one-screen version of this lesson. Switch to Full for the explanations, worked examples and interactives.

Why value more than one future?

A forecast model makes what-if analysis easy: change the story and re-run. In scenario analysis each scenario is a coherent set of assumptions for revenue growth, margins, investment and financing, carried through the statements to its own firm value and equity value. Each one tells its own story, a valuation narrative per scenario . Lesson 7 shows one run in full.

Scenarios can be shown unweighted, as a range (the margin paths in lesson 3), or weighted by probability to give one estimate.

Sensitivity analysis is different. It changes one or more chosen inputs while holding all the others constant, to see how much the value depends on them.

Scenario analysis against sensitivity analysis
Testing a valuation
  • Scenario analysis Memorize a story, with linked assumptions
    • Changes a whole set of linked assumptions, as a story
    • Output an equity value per scenario, then a weighted value
  • Sensitivity analysis Memorize a dial, one input at a time
    • Changes chosen inputs, all others held constant
    • Output how far value moves per input

Exam tip A scenario is a story; a sensitivity is a dial. A “scenario” that changes one number is a sensitivity.

How do you weight scenario values?

Value each scenario, weight each value by its probability and add them up. That gives the probability-weighted equity value .

Probability-weighted equity value Use it
Equity value=∑sps×max⁡(Vs,0),∑sps=1\text{Equity value} = \sum_{s} p_s \times \max(V_s, 0), \qquad \sum_s p_s = 1
InteractiveTessaly Glass: one equity value from four scenarios

Try this 0 of 3

  • Shift the probabilities until the weighted value also says 'undervalued'
  • Turn off the floor and see the weighted value fall
  • On the sensitivity tab, find the input that moves value most
Scenario equity values (CHF m)
Probabilities (always sum to 100%)
Compare

Optimistic 520.0 at 15%, Base 410.0 at 50%, Stable 250.0 at 25%, Downside −30.0 at 10%. Weighted equity value 345.5 against market capitalization 370: overvalued. The base case alone says undervalued.

Equity value today345.5CHF m,probability-weightedOptimistic520.0Base410.0Stable250.0!Downside−30.0counts as 0floored to 0
  • Equity value today, 345.5, CHF m, probability-weighted, (matches)
    • p = 15%: Optimistic, 520.0
    • p = 50%: Base, 410.0
    • p = 25%: Stable, 250.0
    • p = 10%: Downside, −30.0, counts as 0, floored to 0, (warning)

Hover over or tap a box to follow its path.

Weighted value, base case and market capitalization (CHF m)

Weighted 345.5Market cap 370Base 410.0
Probability-weighted equity value345.5
Verdict on the weighted valueovervalued (−6.6%)
Base case alone saysundervalued (+10.8%)
Trap: plain average295.0
Trap: weighted, no floor342.5
Show working

Optimistic: 0.15 × 520.0 = 78.0

Base: 0.5 × 410.0 = 205.0

Stable: 0.25 × 250.0 = 62.5

Downside: 0.1 × 0.0 (computed −30.0, floored to 0) = 0.0

Weighted equity value = 345.5. Against market capitalization of 370: −6.6%, so the shares look overvalued.

Shareholders have limited liability, so a negative equity value counts as zero. Keeping it gives 342.5.

A plain average of the four (520.0, 410.0, 250.0, 0.0) is 295.0: it ignores the probabilities.

Probability-weighted equity value on your calculator
  1. (0.15×520)+(0.5×410)+(0.25×250)+(0.1×0)=345.5
  1. 0.15ENTER520×0.5ENTER410×+0.25ENTER250×+0.1ENTER0×+345.5

ENTER enter (on the HP, puts the number on the stack)

Display: 345.5

A value is a range with weights, and the weights are yours.

Exam tip Probability-weight the equity values, not the inputs. Then floor any negative equity value at zero.

What do the probabilities mean?

  • They are the analyst’s subjective probabilities , a matter of judgment. They are unlikely to equal the true probabilities, which nobody can know.
  • Only one scenario will happen. The weighted value summarizes a range of outcomes; it is not a forecast of any one of them.
  • Scenarios let the analyst build narratives into the value. They do not make the estimate more accurate.
  • The model is revised as quarterly or semiannual reports and other news arrive, and the scenarios and weights with it. That is updating the model .
  • A second weighting tool: for a company that might fail, weight a going-concern value and a distress value by the probability of distress (lesson 6).
Go deeper Why the base case is not the answer

The base case can carry the largest weight and still mislead, because the other scenarios are not symmetric around it. In the Tessaly example the downside and stable cases pull the weighted value 64.5 below the base case.

How does sensitivity analysis look?

Memorize this

MemorizeWeighted scenario value

StepRule
Value each scenariore-run the model to a firm and equity value
Floornegative equity value → 0 (limited liability)
WeightΣ p × value, probabilities sum to 1
Compareweighted value vs market capitalization

Memory hook Run, floor, weight, compare.

Know it cold Probabilities are subjective, not true; scenarios add narrative, not accuracy; update as news arrives.

MemorizeScenario or sensitivity?

Memory hook A scenario is a story; a sensitivity is a dial.

Know it cold Scenario: a coherent set of assumptions changes together. Sensitivity: chosen inputs change, all others held constant.

Scenario analysis against sensitivity analysis
Testing a valuation
  • Scenario analysis Memorize a story, with linked assumptions
    • Changes a whole set of linked assumptions, as a story
    • Output an equity value per scenario, then a weighted value
  • Sensitivity analysis Memorize a dial, one input at a time
    • Changes chosen inputs, all others held constant
    • Output how far value moves per input
Probability-weighted equity value Use it
Equity value=∑sps×max⁡(Vs,0),∑sps=1\text{Equity value} = \sum_{s} p_s \times \max(V_s, 0), \qquad \sum_s p_s = 1
Quick example

Three scenarios: 90 (probability 35%), 60 (40%) and −20 (25%). The negative value counts as zero, so the weighted value is 0.35 × 90 + 0.40 × 60 + 0.25 × 0 = 55.5.

On your calculator
  1. 0.35×90=STO1 31.5
  2. 0.4×60+RCL1= 55.5

STO store · RCL recall a stored number

  1. 0.35ENTER90×0.4ENTER60×+ 55.5

ENTER enter (on the HP, puts the number on the stack)

How it is asked

  1. Weighted value with a negative scenario

    Values 160 (30%), 100 (40%), 60 (25%), −50 (5%).

    48 + 40 + 15 + 0 = 103. Keeping −50 (100.5) is the trap.

  2. Fairly valued?

    Base case 3.9bn, weighted 3.2bn, market cap 3.5bn.

    Judge by the weighted value: about 9% below market cap, so the shares look overvalued. The base case is the trap.

  3. What are the probabilities?

    Are scenario probabilities objective?

    No: subjective; unlikely to equal the true, unknowable probabilities.

  4. Scenario versus sensitivity

    Only terminal growth changes; everything else is held.

    Sensitivity analysis.

Exam traps

  • Comparing the base case, not the weighted value, with market capitalization.
  • Keeping a negative scenario equity value instead of setting it to zero.
  • Weighting the inputs (growth, margins) instead of the scenario values.
  • Treating the probabilities as objective, or the weighted value as more accurate than a single case.
  • Calling a change in one input, with everything else held, a scenario: that is sensitivity analysis.

Checkpoint

3 exam-style questions. About 90 seconds each.

  1. 1Compared with sensitivity analysis, scenario analysis most likely:

  2. 2An analyst values a company's equity under four scenarios: upside CHF 180 million (probability 20%), base CHF 120 million (50%), stable CHF 70 million (20%) and downside −CHF 40 million (10%), where firm value is below the face value of debt. The probability-weighted equity value is closest to:

  3. 3An analyst's base case values a company's equity at USD 2.6 billion. Weighting four scenarios by the analyst's probabilities gives USD 2.1 billion. The company's market capitalization is USD 2.3 billion. The most appropriate conclusion is that the shares appear:

Key takeaways

  • Each scenario re-runs the whole model to its own equity value.
  • Weighted value = Σ probability × equity value, with negative values set to zero.
  • Compare the weighted value, not the base case, with market capitalization.
  • Probabilities are subjective; scenarios add narrative, not accuracy. Sensitivity analysis moves chosen inputs and holds the rest.
Probability-weighted equity value Use it
Equity value=∑sps×max⁡(Vs,0),∑sps=1\text{Equity value} = \sum_{s} p_s \times \max(V_s, 0), \qquad \sum_s p_s = 1
Quick example

Three scenarios: 90 (probability 35%), 60 (40%) and −20 (25%). The negative value counts as zero, so the weighted value is 0.35 × 90 + 0.40 × 60 + 0.25 × 0 = 55.5.

On your calculator
  1. 0.35×90=STO1 31.5
  2. 0.4×60+RCL1= 55.5

STO store · RCL recall a stored number

  1. 0.35ENTER90×0.4ENTER60×+ 55.5

ENTER enter (on the HP, puts the number on the stack)

All Equities formulas, every lesson

Practice exam-style questions on Financial Statement Forecasting

Questions about this lesson?

Where this comes back

Questions candidates ask

How do you calculate a probability-weighted equity value?

Value equity under each scenario, set any negative value to zero, multiply each by its probability and add them up. Compare the result with market capitalization.

Why is a negative scenario equity value set to zero?

Shareholders have limited liability. If firm value falls below debt, the debt's market value falls instead, and equity is worth about zero, not less.

What is the difference between scenario and sensitivity analysis?

A scenario changes a whole set of linked assumptions as one story. Sensitivity analysis changes chosen inputs while holding all others constant.

Spotted an error? Tell us and we will fix it.

  1. Why build a disaggregated model, and at what level of detail
  2. Revenue models: historical, bottom-up and top-down
  3. Expense models: margins, itemization and operating leverage
  4. Asset and financing models
  5. Scenario analysis with probability weights
  6. Model construction by company life cycle: from venture to decline
  7. From the forecast model to an equity value

Ask Sensei Listen instead

Questions on Scenario analysis with probability weights

Learn with Sensei, out loud The voice tutor, in early access

Sensei teaches the lesson aloud and stops when you interrupt: "wait, why is the geometric mean lower?" It answers with the lesson's own example, then quizzes you by voice. Every account can try Sensei on one module, with one voice question.

USD 149 a month founding price (regular USD 199), kept for as long as you stay subscribed. Everything in Plus included.

In active development and improving every day. Send feedback · Plans and pricing

Analyst Dojo Plus

Your free review list holds 10 items.

Plus keeps everything you save, for USD 19 a month. To make room on the free plan, remove something from your review list.

See Plus

Lessons, checkpoints and the lesson questions in practice stay free. Compare plans