Decision trees
Practice Flashcards
Decision trees: □ = decision, ○ = chance, branches = ___
Track your progress — Sign up free to save your progress and get smart review reminders based on spaced repetition.
All Flashcards in Topic 6.6
Below are all 45 flashcards for this topic. Sign up free to track your progress and get personalized review schedules.
6.6.120 cards
Decision trees: □ = decision, ○ = chance, branches = ___
Options and outcomes with probabilities and financial values.
Options + outcomes
What is a decision tree?
A diagram mapping choices, outcomes, probabilities and financial values — helps decide based on expected value, not gut feeling.
Choice diagram + probabilities
Read a decision tree: start from the ___ at the □
Left — follow branches to see options, then outcomes at chance nodes.
Left
Two benefits of decision trees?
Visual (easy to see all options); quantitative (uses numbers); forces risk assessment; compares expected values.
Visual + quantitative + risk
Compare expected values — the highest is usually the best, BUT ___
Always consider qualitative factors and risk tolerance too.
Qual + risk tolerance
Square □ on a decision tree = ___
Decision node — a point where the manager makes a choice.
Decision point
Two limitations of decision trees?
Probabilities are estimates (may be wrong); ignores qualitative factors; complex; gives false precision.
Estimates + no qual + complex
EV = outcome × probability (sum all at each node)
Expected value calculation — multiply and add.
Multiply + add
When evaluating, don't just pick the highest EV — also discuss ___
Risk, stakeholders and non-financial factors.
Risk + stakeholders + non-financial
Circle ○ on a decision tree = ___
Chance node — a point where different outcomes can happen with probabilities.
Chance/outcome point
Benefits: visual, quantitative, forces ___
Risk assessment — managers must think about probabilities.
Risk assessment
Decision trees give a 'false sense of precision' because ___
The calculations look exact but are based on estimated probabilities — garbage in, garbage out.
Estimates look precise
Limitations: estimates, ignores ___, false precision
Qualitative factors (ethics, reputation, morale).
Qualitative
Probabilities at each chance node must add up to ___
1 (or 100%) — if they don't, something is wrong.
1.0
At chance nodes, look at ___ and ___
Probabilities and payoffs — to calculate expected values.
Probabilities + payoffs
Decision trees don't consider ___
Qualitative factors — ethics, reputation, morale, stakeholder impact.
Qualitative factors
A risk-averse business may prefer ___ EV with more ___
Lower; certainty — safer option even if mathematically inferior.
Lower EV + more certainty
Expected value = outcome × ___
Probability — for each outcome, then sum all results at that node.
Probability
Probabilities are ___, so EVs are only as good as ___
Estimates; the estimates used — they may be wrong.
Estimates; estimate quality
Quick: □ = ___. ○ = ___
Decision node; Chance node.
Decision; Chance
6.6.225 cards
Net expected value formula?
Net EV = Expected value − Cost of the option.
EV minus cost
Expected value formula?
EV = Σ (outcome × probability) — add up all outcome×probability pairs at a chance node.
Sum of (outcome × prob)
Draw decision trees ___ to right, calculate ___ to left
Left; right — build the diagram forward, calculate backward.
Draw → , calculate ←
The highest net EV is mathematically best, BUT consider ___
Reliability of probabilities, affordability of losses, stakeholder impact, ethics, risk appetite.
Reliability + loss + stakeholders + ethics
EV = Σ (outcome × probability). Net EV = EV − ___
Cost of the option.
Cost
Draw left to right, calculate ___
Right to left.
Right to left
Step 1: Start with a ___ on the left
Decision node (square □) — the starting choice.
Square □
'How much can the business afford to lose?' relates to ___
Risk tolerance — if failure means bankruptcy, even high EV may be too risky.
Risk tolerance
Option A: success $100k (0.6), failure $20k (0.4). EV?
EV = ($100k × 0.6) + ($20k × 0.4) = $60k + $8k = $68,000.
$68,000
Option A costs $50k, EV = $68k. Net EV?
$68k − $50k = $18,000 net expected gain.
$18,000
In 10-mark questions, use decision tree calculation PLUS ___
Qualitative evaluation — numbers alone aren't enough for top marks.
Qualitative evaluation
Compare ___ EVs to choose between options
Net EVs — accounts for the different costs of each option.
Net
Probabilities at each node must add to ___
1.0.
1.0
Mark rejected options with ___
Two diagonal lines across the branch — shows which option was NOT chosen.
Two diagonal lines
Always check that probabilities at each node add to ___
1.0 — if not, there's an error.
1.0
Why mark rejected branches?
Shows the examiner you understand the decision process — Nov 2024 markscheme awarded marks for this.
Marks for showing rejection
A risk-averse business prefers ___ EV with more ___
Lower; certainty/predictability.
Lower + certain
EV tells you the ___ financial return of each option
Weighted average — not what you'll definitely get, but the statistical expectation.
Weighted average
Always evaluate beyond numbers: discuss ___, ___, qualitative
Risk and stakeholders.
Risk + stakeholders
Highest net EV isn't always best if it's also the ___
Riskiest — a risk-averse business may prefer lower net EV with more certainty.
Riskiest
Show working in EV calculations for ___
Method marks — even if the final number is wrong.
Method marks
Calculate from the ___ side first, then move ___
Right (outcomes); left (to compare at decision node).
Right → left
Net EV shows the expected ___ from choosing that option
Net gain (or loss) — after accounting for the cost of the investment.
Net gain
Quick: Mark rejected branches with ___
Two diagonal lines.
Two lines
Don't just report the EV — also evaluate ___
Whether the probabilities are reliable and what qualitative factors matter.
Reliability + qualitative
Topic 6.6 study notes
Full notes & explanations for Decision trees
BM exam skills
Paper structures, command terms & tips
Want smart review reminders?
Sign up free to track your progress. Our spaced repetition algorithm will tell you exactly which cards to review and when.
Start Free