default
default:
object
Defined in: index.js:72
Default export: every distribution by name, the distributions registry,
createRng, and the special namespace, bundled for a single import.
Type Declaration
bernoulli
bernoulli:
Readonly<{kind:"discrete";name:"bernoulli";params:string[];cdf:number;dlogpdf: {dp:number; };logpdf:number;mean:any;number;quantile:number;sample:0|1;sampleN:any[];support:number[];validate:void;variance:number; }>
Bernoulli distribution, parameterized {p} with success probability p in [0, 1]. Support is {0, 1}.
beta
beta:
Readonly<{kind:"continuous";logpdf: (x,params) =>number;name:"beta";params:string[];cdf:number;dlogpdf: {dalpha:number;dbeta:number;dx:number; };mean:number;number;quantile:number;sample:number;sampleN:any[];support:number[];validate:void;variance:number; }>
Beta distribution, parameterized {alpha, beta} with shape parameters alpha > 0 and beta > 0. Support is the open interval (0, 1).
binomial
binomial:
Readonly<{kind:"discrete";name:"binomial";params:string[];cdf:number;dlogpdf: {dp:number; };logpdf:number;mean:number;number;quantile:any;sample:any;sampleN:any[];support:any[];validate:void;variance:number; }>
Binomial distribution: successes in n independent trials. Parameterized {n, p} with trial count n and success probability p in [0, 1]. Support is the integers {0, 1, …, n}.
chi2
chi2:
Readonly<{kind:"continuous";name:"chi2";params:string[];cdf:number;dlogpdf: {dk:number;dx:number; };logpdf:number;mean:any;number;quantile:number;sample:number;sampleN:any[];support:number[];validate:void;variance:number; }>
Chi-squared distribution, parameterized {k} with degrees of freedom k > 0. Support is (0, infinity).
createRng
createRng: (
seed?) =>any
Create a seedable RNG.
Parameters
seed?
number = ...
Any finite number; omit for a time-based seed
Returns
any
{float, int, normal, seed}
distributions
distributions:
Readonly<{bernoulli:Readonly<{kind:"discrete";name:"bernoulli";params:string[];cdf:number;dlogpdf: {dp:number; };logpdf:number;mean:any;number;quantile:number;sample:0|1;sampleN:any[];support:number[];validate:void;variance:number; }>;beta:Readonly<{kind:"continuous";logpdf: (x,params) =>number;name:"beta";params:string[];cdf:number;dlogpdf: {dalpha:number;dbeta:number;dx:number; };mean:number;number;quantile:number;sample:number;sampleN:any[];support:number[];validate:void;variance:number; }>;binomial:Readonly<{kind:"discrete";name:"binomial";params:string[];cdf:number;dlogpdf: {dp:number; };logpdf:number;mean:number;number;quantile:any;sample:any;sampleN:any[];support:any[];validate:void;variance:number; }>;chi2:Readonly<{kind:"continuous";name:"chi2";params:string[];cdf:number;dlogpdf: {dk:number;dx:number; };logpdf:number;mean:any;number;quantile:number;sample:number;sampleN:any[];support:number[];validate:void;variance:number; }>;exponential:Readonly<{kind:"continuous";logpdf: (x,params) =>number;name:"exponential";params:string[];cdf:number;dlogpdf: {dlambda:number;dx:number; };mean:number;number;quantile:number;sample:number;sampleN:any[];support:number[];validate:void;variance:number; }>;f:Readonly<{kind:"continuous";name:"f";params:string[];cdf:number;dlogpdf: {dd1:number;dd2:number;dx:number; };logpdf:number;mean:number;number;quantile:number;sample:number;sampleN:any[];support:number[];validate:void;variance:number; }>;gamma:Readonly<{kind:"continuous";logpdf: (x,params) =>number;name:"gamma";params:string[];cdf:number;dlogpdf: {dalpha:number;dbeta:number;dx:number; };mean:number;number;quantile:number;sample:number;sampleN:any[];support:number[];validate:void;variance:number; }>;halfnormal:Readonly<{kind:"continuous";logpdf: (x,params) =>number;name:"halfnormal";params:string[];cdf:number;dlogpdf: {dsigma:number;dx:number; };mean:number;number;quantile:number;sample:number;sampleN:any[];support:number[];validate:void;variance:number; }>;lognormal:Readonly<{kind:"continuous";logpdf: (x,params) =>number;name:"lognormal";params:string[];cdf:number;dlogpdf: {dmu:number;dsigma:number;dx:number; };mean:number;number;quantile:number;sample:number;sampleN:any[];support:number[];validate:void;variance:number; }>;normal:Readonly<{kind:"continuous";logpdf: (x,params) =>number;name:"normal";params:string[];cdf:number;dlogpdf: {dmu:number;dsigma:number;dx:number; };mean:any;number;quantile:any;sample:any;sampleN:any[];support:number[];validate:void;variance:number; }>;poisson:Readonly<{kind:"discrete";name:"poisson";params:string[];cdf:number;dlogpdf: {dlambda:number; };logpdf:number;mean:any;number;quantile:number;sample:number;sampleN:any[];support:number[];validate:void;variance:any; }>;studentT:Readonly<{kind:"continuous";name:"studentT";params:string[];cdf:number;dlogpdf: {dmu:number;dnu:number;dsigma:number;dx:number; };logpdf:number;mean:any;number;quantile:any;sample:any;sampleN:any[];support:number[];validate:void;variance:number; }>;uniform:Readonly<{kind:"continuous";logpdf: (x,params) =>number;name:"uniform";params:string[];cdf:number;dlogpdf: {dhigh:number;dlow:number;dx:number; };mean:number;number;quantile:any;sample:any;sampleN:any[];support:any[];validate:void;variance:number; }>; }>
Registry of all distributions keyed by name, for dynamic lookup (e.g. model specifications that name distributions as strings).
exponential
exponential:
Readonly<{kind:"continuous";logpdf: (x,params) =>number;name:"exponential";params:string[];cdf:number;dlogpdf: {dlambda:number;dx:number; };mean:number;number;quantile:number;sample:number;sampleN:any[];support:number[];validate:void;variance:number; }>
Exponential distribution, parameterized {lambda} with rate lambda > 0. Support is [0, infinity).
f
f:
Readonly<{kind:"continuous";name:"f";params:string[];cdf:number;dlogpdf: {dd1:number;dd2:number;dx:number; };logpdf:number;mean:number;number;quantile:number;sample:number;sampleN:any[];support:number[];validate:void;variance:number; }>
F distribution, parameterized {d1, d2} with numerator and denominator degrees of freedom d1 > 0 and d2 > 0. Support is (0, infinity).
gamma
gamma:
Readonly<{kind:"continuous";logpdf: (x,params) =>number;name:"gamma";params:string[];cdf:number;dlogpdf: {dalpha:number;dbeta:number;dx:number; };mean:number;number;quantile:number;sample:number;sampleN:any[];support:number[];validate:void;variance:number; }>
Gamma distribution, parameterized {alpha, beta} with shape alpha > 0 and rate beta > 0. Support is (0, infinity).
halfnormal
halfnormal:
Readonly<{kind:"continuous";logpdf: (x,params) =>number;name:"halfnormal";params:string[];cdf:number;dlogpdf: {dsigma:number;dx:number; };mean:number;number;quantile:number;sample:number;sampleN:any[];support:number[];validate:void;variance:number; }>
Half-normal distribution, the absolute value of a zero-mean normal. Parameterized {sigma} with scale sigma > 0. Support is [0, infinity).
lognormal
lognormal:
Readonly<{kind:"continuous";logpdf: (x,params) =>number;name:"lognormal";params:string[];cdf:number;dlogpdf: {dmu:number;dsigma:number;dx:number; };mean:number;number;quantile:number;sample:number;sampleN:any[];support:number[];validate:void;variance:number; }>
Log-normal distribution: exp(X) where X ~ Normal(mu, sigma). Parameterized {mu, sigma} with sigma > 0. Support is (0, infinity).
normal
normal:
Readonly<{kind:"continuous";logpdf: (x,params) =>number;name:"normal";params:string[];cdf:number;dlogpdf: {dmu:number;dsigma:number;dx:number; };mean:any;number;quantile:any;sample:any;sampleN:any[];support:number[];validate:void;variance:number; }>
Normal (Gaussian) distribution, parameterized {mu, sigma} with mean mu and standard deviation sigma > 0. Support is the whole real line.
poisson
poisson:
Readonly<{kind:"discrete";name:"poisson";params:string[];cdf:number;dlogpdf: {dlambda:number; };logpdf:number;mean:any;number;quantile:number;sample:number;sampleN:any[];support:number[];validate:void;variance:any; }>
Poisson distribution, parameterized {lambda} with rate lambda > 0. Support is the non-negative integers {0, 1, 2, …}.
special
special:
special
Special functions underpinning the distribution numerics: log-gamma, digamma, regularized incomplete gamma/beta and their inverses, erf/erfc, and the inverse normal CDF.
Implemented from the standard published algorithms (Lanczos approximation; series and modified-Lentz continued fractions for the incomplete functions; Acklam’s rational approximation with Halley refinement for the normal quantile). Accuracy targets are 1e-12 or better across the usual parameter ranges; the scipy comparison suite enforces this.
studentT
studentT:
Readonly<{kind:"continuous";name:"studentT";params:string[];cdf:number;dlogpdf: {dmu:number;dnu:number;dsigma:number;dx:number; };logpdf:number;mean:any;number;quantile:any;sample:any;sampleN:any[];support:number[];validate:void;variance:number; }>
Student’s t distribution, parameterized {nu} with degrees of freedom nu > 0. Support is the whole real line.
uniform
uniform:
Readonly<{kind:"continuous";logpdf: (x,params) =>number;name:"uniform";params:string[];cdf:number;dlogpdf: {dhigh:number;dlow:number;dx:number; };mean:number;number;quantile:any;sample:any;sampleN:any[];support:any[];validate:void;variance:number; }>
Continuous uniform distribution, parameterized {low, high} with low < high. Support is the interval [low, high].