Built-in distributions
Understand D/P/Q/R and answer probability questions with familiar laws.
Tutorial →Response-time vignette →
Learning center
Learn probability modeling by doing
Begin with a single distribution, graduate to composition and dependence, then fit Bayesian models with MCMC. Tutorials teach the API; companion vignettes work through decisions and computational review.
Choose a path
Seven paths, no prerequisites beyond Java
Response-time vignette →
Compose and transform
Mix populations, restrict support, represent measurement limits, change units, and apply a Jacobian correctly.
Tutorial →Custom distributions
Convert a nonnegative kernel or set of weights into a complete numerical distribution.
Tutorial →Sensor-error vignette →
Copulas and dependence
Keep marginal behavior separate from dependence, including mixed and vine models.
Tutorial →Claims vignette →
Bayesian models and MCMC
Use one organized gateway to choose a complete quick start, pure-Java modeling, feature examples, JDM/Stan scripts, beginner language material, or advanced diagnostics.
Choose your MCMC route →Tails and option-implied laws
Move from loss/return conventions through heavy tails, EVT, curve repair, and labeled Q or physical predictive outputs.
Finance tutorial →Worked options analysis →
Dense and sparse linear algebra
Choose CPU or GPU providers, reuse dense and sparse operands, solve systems, and factor matrices through matching FP64 and FP32 contracts.
Linear-algebra guide →Complete API contract →
After hypothesis tests · 0.7.0+
Control errors across many p-values
The multiple-testing guide covers Bonferroni, Holm, Hochberg, Hommel, Šidák, BH, BY, adaptive BKY and GBS, weighted FWER/FDR, grouped and online procedures, discrete DBH, log-scale and censored inputs, and Storey q-values through one compact static API.
Keep nearby
References for the details
The distribution catalog answers “which law and parameterization?”, while the JavaDoc lists every overload. The tutorials deliberately focus on the small set beginners need.