Built-in aggregate functions

Aggregate Function	Argument(s)	Returns	Result Description
avg(value)	decimal or float	decimal or float	The arithmetic mean of the values in the window. The avg() and mean() functions do exactly the same thing. They are aliases for each other.
count()	-	integer	The number of items in the window, including any NaN items
count (predicate)	boolean	integer	The number of items for which the argument is true
count(value)	decimal or float	integer	The number of items where the decimal or float value is not NaN
first(value)	decimal or float	decimal or float	The earliest value in the window being aggregated over
last(value)	decimal or float	decimal or float	The latest value in the window being aggregated over
max(value)	decimal or float	decimal or float	The maximum value
mean(value)	decimal or float	decimal or float	The arithmetic mean of the values in the window. The mean() and avg() functions do exactly the same thing. They are aliases for each other
min(value)	decimal or float	decimal or float	The minimum value
nth(value,index) nth(value,0) returns the same item as first(value).	decimal, integer or float, integer	decimal or float	The value of the specified decimal or float item in the index position, starting with the earliest item in the window (item 0) and moving toward the latest item.
prior(value,index) prior(value,0) returns the same item as last(value).	decimal, integer or float, integer	decimal or float	The value of the specified decimal or float item in the index position, starting with the most recent item in the window (item 0 ) and moving toward the earliest item.
stddev(value)	decimal or float	decimal or float	The standard deviation of the values
stddev2(value)	decimal or float	decimal or float	The sample standard deviation of the values
sum(value)	decimal, float or integer	decimal, float or integer	The sum of the values
wavg(value,weight)	decimal,decimal or float,float	decimal or float	The weighted average of the values where each value is weighted by the corresponding weight

Calculations by the built-in aggregate functions might be affected by underflow and overflow. For example, adding a very large number to the collection that the sum() function operates on, then adding a very small number, and then removing the very large number will probably result in 0, and not the very small number. Just adding the very small number would result in behavior that you would expect. The overflow and underflow characteristics are as defined for IEEE 64-bit floating point numbers.