simpm.dist — SimPM’s Probability Distributions components

Probability distributions and fitting helpers used throughout SimPM.

The functions in this module wrap common SciPy distributions with a lightweight API tailored for simulation modeling. They make it easy to sample, plot, and fit distributions while keeping consistent return types for downstream components.

simpm.dist.fit(data: Iterable[float], dist_type: str, method: str = 'mle') → FitResult

Fit a distribution on data and return a structured result.

Parameters

dataarray_like

The data to fit the distribution to.

dist_typestr

The type of distribution to fit. Supported types are "triang", "norm", "beta", "trapz", and "expon".

methodstr, optional

The method to use for fitting the distribution. Default is "mle" (maximum likelihood estimation).

Returns

FitResult

Structured fitting result containing the SciPy distribution, wrapper instance, and Kolmogorov–Smirnov statistic.

Raises

ValueError

If the distribution type is not recognized or the input data are invalid.

class simpm.dist.beta(a, b, minp, maxp)

Defines a beta distribution.

class simpm.dist.distribution

Lightweight wrapper around SciPy distributions.

All concrete distribution classes inherit from this base to expose a common API for sampling, plotting, and computing summary statistics.

cdf(x)

Evaluate the cumulative distribution function at x.

cdf_xy(size: int)

Return x/y pairs for plotting the cumulative distribution function.

mean()

Return the distribution mean.

pdf(x)

Evaluate the probability density function at x.

pdf_xy(size: int = 100)

Return x/y pairs for plotting the probability density function.

percentile(q)

Return the value at percentile q (0–100).

plot_cdf()

Plot the cumulative distribution function using matplotlib.

plot_pdf()

Plot the probability density function using matplotlib.

sample()

Draw a single random variate from the distribution.

samples(n)

Draw n random variates from the distribution.

std()

Return the distribution standard deviation.

var()

Return the distribution variance.

class simpm.dist.empirical(data)

Defines an empirical distribution based on observed data.

cdf(x)

Evaluates the cumulative distribution function (CDF) of the empirical distribution at the given point.

cdf_xy()

Returns x, y numpy arrays for plotting the cumulative distribution function (CDF).

pdf(x)

Evaluates the probability density function (PDF) of the empirical distribution at the given point.

pdf_xy()

Returns x, y numpy arrays for plotting the probability density function (PDF).

percentile(q)

Calculates the value corresponding to the given percentile of the empirical distribution.

plot_cdf()

Plots the cumulative distribution function (CDF) of the empirical distribution.

plot_pdf()

Plots the probability density function (PDF) of the empirical distribution.

sample()

Generates a single random sample from the empirical distribution.

samples(n)

Generates random samples from the empirical distribution.

class simpm.dist.expon(mean)

Defines an exponential distribution.

class simpm.dist.norm(mean, std)

Defines a normal distribution.

class simpm.dist.trapz(a, b, c, d)

Defines a trapezoidal distribution.

class simpm.dist.triang(a, b, c)

Defines a triangular distribution.

class simpm.dist.uniform(a, b)

Uniform distribution defined by lower/upper bounds.