correlation builtin

correlation(
    data, *, columns=None, weights=None, method="pearson"
)

Correlation matrix of the columns of data, each pair over rows without NaN.

Parameters:

Name	Type	Description	Default
data	(array_like, PointSet, BlockModel, Table or dict)	(n, d) values, or named columns.	required
columns	sequence of str	Columns of data to correlate, in order; default all of them.	None
weights	array_like or str	Declustering weights, or their column in data.	None
method	(pearson, spearman, covariance)	Pearson correlation, rank correlation, or the covariance (over the sum of weights, as the variance of describe), variances on the diagonal.	"pearson"

Returns:

Type	Description
ndarray	(d, d) symmetric matrix.

used in

• correlations
• compositional data
• variogram sets
• Iron ore: several grades, one closure, reconciled