Module robust 

Robust statistics: low sensitivity to outliers, useful when the raw distribution has a heavy tail (which latency typically does). Behind the robust Cargo feature (on by default).

Functions

coefficient_of_variation

Coefficient of variation: stddev / mean. Unit-free measure of relative variability. 0.0 for empty / zero-mean input.

iqr

Interquartile range (p75 - p25). The “middle 50%” spread. More robust to outliers than stddev.

kurtosis

Excess kurtosis (4th standardised moment minus 3). Positive excess kurtosis means heavier tails than a normal distribution (typical for latency). Returns 0 with fewer than 4 samples.

median_absolute_deviation

Median absolute deviation: the median of |x - median(x)|. Robust alternative to stddev. Returns 0 for empty input.

skewness

Skewness (3rd standardised moment). Positive skew means a right tail (typical for latency distributions). Returns 0 with fewer than 3 samples.