Sx is the support of an event, AKA the set of outcomes that can occur

  • For example, when flipping a coin, the support can be thought of as 0 and 1, where 0 is heads and 1 is tails

f(x) = P(X = x) corresponds to a pmf (probability mass function)

  • Accessed in R using d

F(x) = P(X ≤ x) corresponds to a cdf (continuous discrete function)

  • Accessed in R using p

E(X) corresponds to the Expectation of X (or expected value)

VAR(X) corresponds to the variance of X, where the standard deviation of a set is equal to the square root of the variance.

  • VAR(X) = E(X2) - E(X)2
  • SD = sqrt(VAR(X))