Module stats 

Rolling statistical primitives, linear regression, and sample statistics (Wilson interval, longest run). Rolling statistics primitives for streaming series calculations.

Structs

LinearRegressionResult

Result of a linear regression fit over a data slice.

ProportionInterval

A binomial proportion with its confidence bounds, all in 0.0..=1.0.

Functions

correlation

Computes Pearson correlation from finite, positionally aligned pairs.

linear_regression

Computes ordinary least squares (OLS) linear regression over a slice of values (where X is 0..N-1).

longest_run

Length of the longest unbroken run of consecutive elements for which predicate holds; 0 if it never does.

percent_rank

Returns the percentile rank (0.0..=100.0) of val within slice.

rolling_mean

Returns the arithmetic mean of all finite values in slice.

rolling_median

Returns the median of finite values in slice.

rolling_quantile

Returns the quantile (0.0..=1.0) of finite values in slice.

rolling_stddev

Returns the standard deviation of finite values in slice.

rolling_sum

Returns the sum of all finite values in slice.

rolling_variance

Returns the population variance of finite values in slice.

wilson_interval

Wilson score interval for successes out of trials.