pub fn mean(data: &[f64]) -> Option<f64>Expand description
Computes the arithmetic mean using Kahan compensated summation.
§Algorithm
Kahan summation accumulates a compensation term to recover lost
low-order bits, achieving O(ε) total error independent of n.
§Complexity
Time: O(n), Space: O(1)
§Returns
Noneifdatais empty or contains any NaN/Inf.
§Examples
use u_numflow::stats::mean;
let v = [1.0, 2.0, 3.0, 4.0, 5.0];
assert!((mean(&v).unwrap() - 3.0).abs() < 1e-15);