kurtSamp

Introduced in: v20.1

Computes the sample kurtosis of a sequence.

It represents an unbiased estimate of the kurtosis of a random variable if passed values form its sample.

Syntax

kurtSamp(expr)

Arguments

expr — Expression returning a number. (U)Int* or Float* or Decimal

Returned value

Returns the kurtosis of the given distribution. If n <= 1 (n is a size of the sample), then the function returns nan. Float64

Examples

Computing sample kurtosis

CREATE TABLE test_data (x Float64) ENGINE = Memory;
INSERT INTO test_data VALUES (1), (2), (3), (4), (5), (6), (7), (8), (9), (10);

SELECT kurtSamp(x) FROM test_data;

┌────────kurtSamp(x)─┐
│ 1.4383636363636365 │
└────────────────────┘

kurtSamp​

kurtSamp