Lesson 3 of 6 · All lessons in this module
  1. Expected value, variance and standard deviationVideo lesson
  2. Conditional probability and probability trees
  3. Conditional expected values and variancesVideo lesson
  4. Bayes' formula: updating a probability with new informationVideo lesson
  5. Bayes in practice: diffuse priors, missing inputs and screening tests
  6. Integrated cases: a tree with model forecasts and a Bayes update

Lesson 3 of 6 · 15 min

Conditional expected values and variances

Work out the expected value inside each scenario, then weight those conditional expectations by the scenario probabilities: the result must match the overall forecast.

In short

  • A conditional expected value E(X∣S)E(X \mid S) averages the outcomes using their probabilities given scenario S.
  • The total probability rule for expected value: E(X)=∑E(X∣Si)P(Si)E(X) = \sum E(X \mid S_i)P(S_i).
  • Conditional and unconditional forecasts must be consistent; if they are not, other investors can profit at your expense.
  • Each scenario has its own conditional variance, measured around E(X∣S)E(X \mid S), which shows the risk within that scenario.
  • When a scenario is revealed, the best forecast moves from E(X)E(X) to E(X∣S)E(X \mid S). This is how analysts update.

Unlock this lesson free for 7 days

Create a free account to get 7 days free: every lesson, video and flashcard, and the question bank. No card needed.

Start your 7-day free trialLog in