COUNT

Record count

COUNT_NULLS

Count of null values

COUNT_NOT_NULLS

Count of not null values

DISTINCT

Distinct count (not Null)

UNIQUE

Unique count (not Null)

SUM

Summation

AVG

Average

MEDIAN

Median (50th Percentile)

STD

Standard Deviation (for more information, see: MATLAB Reference (2))

VAR

Variance (for more information, see: MATLAB Reference)

FIRST_X

First X values - lower extreme

LAST_X

Last X values - upper extreme

MIN_LENGTH

Minimum length of sequence

MIN_POSITIVE_LENGTH

Minimum length of non-empty sequence

AVG_LENGTH

Average length of sequence

MEDIAN_LENGTH

Median length of sequence

MAX_LENGTH

Maximum length of sequence

QUANTILES

Dividing sorted data (excluding null values) into roughly equal intervals based on the parameter value, count + 1, and where the interval length is L = (N / (C)), such that N is element count and C is quantile count. The quantiles are the data values marking the boundaries between consecutive intervals/data subsets. Given 100 records and a value of parameter count = 4, it follows that L = (100 / (4)). Thus L = 20, and interval positions will be at the following positions: 0, 20, 40, 60, 80, 99 of the sorted records counted from zero. When the element set cannot be divided into a whole number given the specified number of quantiles, no interpolation is performed. Instead, L is a real number and the positions are calculated by rounding down to whole numbers. Example: Consider 100 given names where the value of the 20% quantile is "Jane". This implies that when all names are sorted alphabetically (locale specific), then 20% of all records occur before the sorted position of "Jane" (area less than and equal to), and 80% of all records occur after the sorted position of "Jane" (greater than and equal to). This applies even when all records would contain the given name "Jane".