Aggregate functions calculate a single value for a collection of
    rows in a result table. Aggregate functions are sometimes called
    statistical functions.

    The following notes generally apply to aggregate functions. An
    aggregate function is a single value derived from one or more
    sets of values.

    o  A value expression is used to evaluate a value for each row.
       The aggregate function operates on these values.

    o  Null values are not included when SQL evaluates functions. If
       you specify DISTINCT, redundant values are also not included.
       If you have set the dialect to SQL99, this null elimination
       causes a warning to be returned for the SQLCODE or SQLSTATE.
       See the Oracle Rdb SQL Reference Manual for more information
       on SQLSTATE and SQLCODE.

    o  If a function has as its argument a value expression that
       contains a column name that is an outer reference (see Outer_
       References), the value expression cannot include an arithmetic
       operator. (The only cases where an outer reference makes sense
       as the argument to a function is in the subquery of a HAVING
       clause or in a subquery in a select list.)

    o  You cannot nest functions. This means that a value expression
       used as an argument to a function cannot include a function.

    o  The keyword ALL in SUM, AVG, MAX, and MIN has no effect. For
       instance, specifying MAX (ALL EMPLOYEE_ID) is the same as
       saying MAX (EMPLOYEE_ID).

    o  VARIANCE and STDDEV must be passed a single numeric value
       expression and the result is returned as a DOUBLE PRECISION
       value. Use the CAST function to alter the result data type.

    o  VARIANCE, and hence STDDEV, assume that one degree of freedom
       is used in the calculation of the mean (average) and therefore
       the divisor is specified as (n-1). For a large number of
       values in the statistical sample this will not be significant.
       However, for small samples it may not be desirable, so this
       default can be changed for the current session by using the
       SET FLAGS option VARIANCE_DOF(0). Only the values 0 and 1 are
       allowed.

    o  The keywords ALL and DISTINCT are not permitted when using
       the VAR_POP, VAR_SAMP, STDDEV_POP and STDDEV_SAMP statistical
       functions.

    The FILTER clause is provided for all statistical functions. This
    clause can be used to limit the values included in the COUNT,
    MAX, MIN, SUM, AVG, STDDEV, and VARIANCE functions.

    FILTER can be used to eliminate data from the statistical
    function so that the generated report can process the data in
    a single pass. The following example illustrates how the FILTER
    clause is applied.

    SQL> select
    cont>   max (salary_amount) filter (where salary_end is null),
    cont>   max (salary_amount) filter (where salary_end is not null),
    cont>   min (distinct salary_amount) filter (where salary_end = salary_start),
    cont>   min (distinct salary_amount) filter (where salary_end > salary_start)
    cont> from
    cont>   salary_history
    cont> where
    cont>   employee_id = '00164'
    cont> group by
    cont>   employee_id;

          51712.00        50000.00            NULL        26291.00
    1 row selected
    SQL>