Independence
Two events are independent when knowing one tells you nothing about the other. Then the chance of both is just the product of their chances.
01The formula
- the chance that both A and B happen
- the chance of each one on its own
- the chance of A once you know B happened
02See it move
03Where it is used
- Risk managerTests whether losses on two desks or two counterparties really are unrelated before multiplying their default chances together.
- Quant researcherChecks whether today's return tells you anything about tomorrow's; if they are independent, past prices carry no signal.
- Execution traderTreats fills on separate venues as independent only when no common news is driving them.
04Why traders use it
Multiplying probabilities is the most common shortcut in risk, and it is only valid under independence. In a crash, assets that looked independent start falling together, and the product badly understates the chance of a joint loss.
05Worked example
Stock A rises with chance 50% and stock B with chance 60%. With no dependence, both rise 30.0% of the time, exactly 30.0%, and the chance A rises is 50.0% whether B rose or not.
06How it is built
- Start from the multiplication rule: .
- If knowing does not change the chance of , then .
- Substitute: .
- To test it with data, compare the share of days when both happened with the product of the two separate shares.
07The common mistake
Confusing independent with mutually exclusive. Events that cannot happen together are strongly dependent: if one happens, you know the other did not.
08In Python and LaTeX
pA, pB = 0.5, 0.6
both = pA * pB # independent
print('P(A and B) =', both, ' P(A|B) =', both / pB) # 0.3, 0.5A, B \text{ independent} \iff P(A \cap B) = P(A)\,P(B) \iff P(A \mid B) = P(A)09Where it fits
Rather not code the maths yourself? Describe your trading idea and let AI build the bot.
Build a bot with AgenKitThe worked example and the model run the same code, so every number matches. Models simplify markets on purpose. This is education, not financial advice.