Irurueta Statistics

This documentation was generated with the assistance of AI. Please report any inaccuracies.

Irurueta Statistics is a lightweight Java library for generating pseudo-random values and for working with common statistical distributions. It provides randomizers that produce uniformly or Gaussian distributed values, distribution classes that evaluate probability density, cumulative distribution, and inverse cumulative distribution functions, and low level special functions used to implement them.

The library is useful when an application needs repeatable pseudo-random data for testing or simulation, needs to evaluate how likely a value is under a normal or chi-squared distribution, or needs to propagate Gaussian uncertainty through a non-linear function.

What the library contains

The public API is organized around three concepts:

graph TD Randomizer["Randomizer base class"] Uniform["UniformRandomizer"] Gaussian["GaussianRandomizer"] Dist["Distribution classes"] Normal["NormalDist"] ChiSq["ChiSqDist"] Special["Special functions"] Erf["Erf"] Gamma["Gamma"] Randomizer --> Uniform Randomizer --> Gaussian Dist --> Normal Dist --> ChiSq Special --> Erf Special --> Gamma Normal --> Erf ChiSq --> Gamma

Randomizer: Abstract base class for pseudo-random value generation, backed by a java.util.Random (or java.security.SecureRandom) instance. Concrete subclasses are UniformRandomizer and GaussianRandomizer, which generate booleans, integers, longs, floats, doubles, and arrays of each following a uniform or Gaussian distribution.
NormalDist: Evaluates the probability density function, cumulative distribution function, and inverse cumulative distribution function of a Gaussian (normal) distribution, computes Mahalanobis distance, and propagates Gaussian uncertainty through a differentiable one-dimensional function.
ChiSqDist: Evaluates the probability density function, cumulative distribution function, and inverse cumulative distribution function of a chi-squared distribution with a given number of degrees of freedom.
Erf and Gamma: Static utility classes implementing the error function and the gamma function, used internally by NormalDist and ChiSqDist and available for direct use in other numerical computations.

Quick example

import com.irurueta.statistics.GaussianRandomizer;
import com.irurueta.statistics.NormalDist;
import com.irurueta.statistics.UniformRandomizer;

UniformRandomizer uniformRandomizer = new UniformRandomizer();
double angle = uniformRandomizer.nextDouble(0.0, 2.0 * Math.PI);

GaussianRandomizer gaussianRandomizer = new GaussianRandomizer(0.0, 1.0);
double noise = gaussianRandomizer.nextDouble();

NormalDist normalDist = new NormalDist(0.0, 1.0);
double probabilityBelowOne = normalDist.cdf(1.0);
double valueAt95Percent = normalDist.invcdf(0.95);

Documentation map

Start with Installation to add the dependency. Then read Random Value Generation for pseudo-random value generation and Statistical Distributions for probability distributions and uncertainty propagation.