Distributions
Distributions provides MATLAB classes for modeling probability distributions and stochastic processes.
These classes are core to the robust fitting techniques in Spline Core.
Quick start
The abstract Distribution class is used through concrete subclasses such as NormalDistribution.
distribution = NormalDistribution();
z = distribution.rand([1000 1]);
distribution.rho = @(tau) exp(-(tau/10).^2);
signal = distribution.noise((0:500).');
Distribution subclasses provide pdf, cdf, total variance, distribution-fit metrics, and correlated-noise generation through rho and noise.
The package also ships runnable demos in the repository Examples folder and formal automated coverage in UnitTests/DistributionsUnitTests.m.
NetCDF persistence
Annotated persistence lets you write a distribution to a NetCDF file and reconstruct it later from either the abstract Distribution role or the concrete class.
distribution = StudentTDistribution(nu=4.5, sigma=1.0);
distribution.writeToFile("student-t.nc", shouldOverwriteExisting=true);
distributionBack = Distribution.distributionFromFile("student-t.nc");