Number of Random Variables

  • univariate probability distribution - sample space is a univariate random variable (one-dimensional variable)
  • joint probability distribution (compound probability distribution) - refers to the probability distribution of 2 or more random variables occurring together
    • multivariate joint probability distribution - sample space is a multivariate random variable (2 or more dimensions/variables) (e.g. the probability of 𝐴 and 𝐵 and 𝐶, denoted 𝐏(𝐴,𝐵,𝐶))
      • bivariate joint probability distribution - sample space is a bivariate random variable (two-dimensional variables) (e.g. the probability of 𝐴 and 𝐵, denoted 𝐏(𝐴,𝐵))
    • full joint probability distribution - refers to a probability distribution of all random variables in a given domain

the following imply each other:

  • 𝐏(𝐴,𝐵) = 𝐏(𝐵,𝐴) = 𝐏(𝐵|𝐴)𝐏(𝐴) = 𝐏(𝐴|𝐵)𝐏(𝐵) # product rule
  • 𝐏(𝐵|𝐴)𝐏(𝐴) = 𝐏(𝐴|𝐵)𝐏(𝐵)/𝐏(𝐴) # Baye’s Theorem