The weighted average of some function under a probability distribution is called the expectation or expected value of and defined by .
For a discrete distribution, this is given by summing over all possible values of in the form
where the average is weighted by the relative probabilities of the different values of .
For continuous distributions, expectations are expressed in terms of integration with respect to the corresponding Probability Density Function:
Approximation from Sample
If we are given a finite number of points drawn from the probability distribution/density, then the expectation can be approximated as a finite sum over these points:
This approximation becomes exact in the limit as .
Multivariate Expectation
For functions of several variables, we can use a subscript to indicate which variable is being averaged over, so that for instance
denotes the average of the function with respect to the distribution of . Note that will be a function of .
Conditional Distribution
We can also consider a conditional expectation with respect to a conditional distribution, so that
which is a function of .
For continuous variables, this conditional expression becomes