The indexed distribution is a general format where both probability and random variable values depend on an integer index. As for the User distribution the probability values need not sum to 1, but are converted to a valid probability by dividing by the sum.

Example: The same example used for the User distribution is shown at the left. There are two separate arrays, one for probabilities and one for values. They are linked by an index.

The indexed random variable is particularly useful when a problem involves several random variables with a similar number of possibilities but different values.