API Reference¶

Import the public API from mcerp:

from mcerp import *

For explicit imports:

from mcerp import N, U, Exp, Gamma, Beta, correlate, correlation_matrix

Core Classes¶

Name	Purpose
`UncertainVariable`	An input uncertainty backed by a `scipy.stats` distribution.
`UncertainFunction`	A calculated uncertain value produced by arithmetic or functions.
`uv(rv, tag=None)`	Create an uncertain variable directly from a frozen `scipy.stats` distribution.
`npts`	Global sample count. Default is `10000`. Set before creating variables.

Tracking Variables¶

Constructors accept an optional tag=... keyword argument. This is useful when several variables share the same distribution but represent different physical quantities:

from mcerp import N

length = N(24, 1, tag="length")
width = N(24, 1, tag="width")

length.tag
width.describe()

If .describe() is called without an explicit name, the tag is shown in the printed output.

Common Properties and Methods¶

Attribute or method	Description
`.mean`	Sample mean.
`.var`	Sample variance.
`.std`	Sample standard deviation.
`.skew`	Standardized skewness coefficient.
`.kurt`	Standardized kurtosis coefficient.
`.stats`	`[mean, var, skew, kurt]`.
`.percentile(p)`	Value at percentile `p`, where `p` is between `0` and `1`.
`.describe(name=None)`	Print a formatted moment summary.
`.plot(hist=False, show=False, **kwargs)`	Plot a distribution or sampled output. Requires Matplotlib.
`.show()`	Display pending Matplotlib plots.

Continuous Distributions¶

Constructor	Distribution
`Beta(alpha, beta, [low, high])`	Beta distribution
`BetaPrime(alpha, beta, [loc, scale])`	Beta prime distribution
`Bradford(q, [low, high])`	Bradford distribution
`Burr(c, k)`	Burr distribution
`ChiSquared(k)` or `Chi2(k)`	Chi-squared distribution
`Erf(h)`	Error-function distribution
`Erlang(k, lamda)`	Erlang distribution
`Exponential(lamda)` or `Exp(lamda)`	Exponential distribution
`ExtValueMax(mu, sigma)` or `EVMax(mu, sigma)`	Extreme value maximum distribution
`ExtValueMin(mu, sigma)` or `EVMin(mu, sigma)`	Extreme value minimum distribution
`Fisher(d1, d2)` or `F(d1, d2)`	F distribution
`Gamma(k, theta)`	Gamma distribution
`LogNormal(mu, sigma)` or `LogN(mu, sigma)`	Log-normal distribution
`Normal(mu, sigma)` or `N(mu, sigma)`	Normal distribution
`Pareto(q, a)`	Pareto distribution, first kind
`Pareto2(q, b)`	Pareto distribution, second kind
`PERT(low, peak, high)`	PERT distribution
`StudentT(v)` or `T(v)`	Student's t distribution
`Triangular(low, peak, high)` or `Tri(low, peak, high)`	Triangular distribution
`Uniform(low, high)` or `U(low, high)`	Uniform distribution
`Weibull(lamda, k)` or `Weib(lamda, k)`	Weibull distribution

Example:

from mcerp import N, Uniform, Gamma

x = N(10, 1)
y = Uniform(0, 5)
z = Gamma(9, 1 / 6)

Discrete Distributions¶

Constructor	Distribution
`Bernoulli(p)` or `Bern(p)`	Bernoulli distribution
`Binomial(n, p)` or `B(n, p)`	Binomial distribution
`Geometric(p)` or `G(p)`	Geometric distribution
`Hypergeometric(N, n, K)` or `H(N, n, K)`	Hypergeometric distribution
`Poisson(lamda)` or `Pois(lamda)`	Poisson distribution

Example:

from mcerp import H, Pois

h = H(50, 5, 10)
p = Pois(3)

SciPy Distributions¶

Any compatible frozen scipy.stats distribution can be wrapped with uv:

import scipy.stats as ss
from mcerp import uv

x = uv(ss.norm(loc=10, scale=1))

Correlation Utilities¶

Function	Description
`correlate(params, corrmat)`	Reorder samples in-place to approximately match a target correlation matrix.
`correlation_matrix(values)`	Return the sample correlation matrix.
`covariance_matrix(values)`	Return the sample covariance matrix.
`plotcorr(values, labels=None, full=False, show=False)`	Plot pairwise sample relationships. Requires Matplotlib.
`induce_correlations(data, corrmat)`	Return correlated sample data for a target matrix.
`chol(matrix)`	Cholesky-like helper used by correlation routines.

Mathematical Functions¶

Use mcerp.umath when applying mathematical functions to uncertain values:

import mcerp.umath as umath

y = umath.sqrt(x) + umath.sin(x)

Available functions include:

acos, acosh, asin, asinh, atan, atanh, ceil, cos, cosh, degrees, exp, expm1, fabs, floor, hypot, ln, log, log10, log1p, radians, sin, sinh, sqrt, tan, tanh, and trunc.