Math
Distributions, statistics, linear algebra, optimization, root finding and Monte Carlo.
The numerical foundation the rest of the library uses, exposed for your own models: normal and other distributions, descriptive statistics and time-series tests, matrix decompositions, optimizers, root finders, interpolation, Kalman filtering and quasi-random sequences.
Common APIs
fn
normalCdfStandard normal cumulative distribution function Φ(x).fnnormalInverseCdfInverse standard normal CDF (quantile function).fnquantileQuantile via the type-7 (linear interpolation of order statistics) method — the default used by NumPy and R.fncorrelationPearson correlation coefficient.fncholeskyCholesky factorization: lower-triangular `L` with `A = L Lᵀ`.fnnearestCorrelationA valid correlation matrix near `matrix`: ONE eigenvalue-clipping pass to PSD ({@link nearestPsd}), then a rescale to a unit diagonal.fnfindRootSafe hybrid solver: expand a bracket if needed, then run Brent.fnbrentBrent's method: inverse-quadratic / secant interpolation with a bisection safeguard.fnnelderMeadNelder–Mead simplex minimization (derivative-free).fnmakePchipInterpolatorBuild a PCHIP (monotone cubic Hermite) interpolator — shape-preserving, no overshoot.fnkalmanFilterRun the forward Kalman recursion over `observations`.fnmonteCarloSee the reference for its full contract.Complete API reference
Browse all public exports, types, and feature imports below.