covarSamp

Introduced in: v1.1

Calculates the sample covariance:

\frac{\Sigma{(x - \bar{x})(y - \bar{y})}}{n - 1}

Примечание

Эта функция использует численно нестабильный алгоритм. Если вам нужна численная стабильность в вычислениях, используйте функцию covarSampStable. It works slower but provides a lower computational error.

Syntax

covarSamp(x, y)

Aliases: COVAR_SAMP

Arguments

x — First variable. (U)Int* or Float* or Decimal
y — Second variable. (U)Int* or Float* or Decimal

Returned value

Returns the sample covariance between x and y. For n <= 1, nan is returned. Float64

Examples

Basic sample covariance calculation

DROP TABLE IF EXISTS series;
CREATE TABLE series(i UInt32, x_value Float64, y_value Float64) ENGINE = Memory;
INSERT INTO series(i, x_value, y_value) VALUES (1, 5.6,-4.4),(2, -9.6,3),(3, -1.3,-4),(4, 5.3,9.7),(5, 4.4,0.037),(6, -8.6,-7.8),(7, 5.1,9.3),(8, 7.9,-3.6),(9, -8.2,0.62),(10, -3,7.3);

SELECT covarSamp(x_value, y_value)
FROM series

┌─covarSamp(x_value, y_value)─┐
│           7.206275555555556 │
└─────────────────────────────┘

Single value returns NaN

SELECT covarSamp(x_value, y_value)
FROM series LIMIT 1

┌─covarSamp(x_value, y_value)─┐
│                         nan │
└─────────────────────────────┘

covarSamp​

covarSamp