Skip to main content

Probability and statistics · 13 min read

Conditional expectation

A forecast is an average given what you know. The tower property, the best predictor, and how much of a return a signal can explain.

Before you start

  • Expected value and variance
  • How sure is an average? (this track)

By the end you'll be able to

  • Compute expectations conditional on events and on random variables
  • Estimate E[Y | X] from data by slicing and averaging, and judge the trade-off
  • Solve branching problems with the tower property
  • Split variance into the part a signal explains and the rest

A trading signal is a forecast, and a forecast is a conditional expectation: the average outcome given what you know. Given that buyers have outnumbered sellers for the last minute, what return should you expect over the next one? This lesson builds the idea from a die roll up to the two identities that every quant uses without thinking: the tower property and the law of total variance.

SymbolMeaning
E[Y]\mathbb{E}[Y]The expected value (long-run average) of YY
E[Y∣A]\mathbb{E}[Y \mid A]The expected value of YY given that event AA happened
E[Y∣X]\mathbb{E}[Y \mid X]The conditional expectation of YY given XX, a function of XX
m(x)m(x)That function: m(x)=E[Y∣X=x]m(x) = \mathbb{E}[Y \mid X = x]
Var⁡(Y)\operatorname{Var}(Y)The variance of YY

The rest of this lesson needs lifetime access

It will be part of lifetime access, which is coming soon: every lesson in all four tracks, including lessons added later, for one payment.

See pricing

Already have access? Sign in