Module joint 

Distributions whose outcomes own continuous draws: what simulate returns when its value holds outcomes of draws made inside it (docs/semantics.md, section 8).

Each outcome’s draws are numbered apart from any world’s, from OWNED, in the order of the latents they were. Drawing from the distribution gives them fresh numbers in the world, so that two draws are independent and the parts of one outcome, like the fields of a record, stay correlated. When sampling, an outcome is chosen, then a value for each of its draws.

Constants

OWNED

Where an outcome’s own draws are numbered from: above any world’s.

Functions

binary

op between a joint distribution and a value that isn’t one, outcome by outcome: joint again, or an ordinary distribution of what’s left, like the truth values of comparisons.

fresh

An outcome drawn into a world: its draws with fresh numbers after next.

mixture

The marginal of a joint distribution of numbers: the mixture of its outcomes’. None if an outcome isn’t a number.

of

The distribution of values that may hold outcomes of draws made where they were computed, as simulate gives it: each outcome’s draws its own, and outcomes of only events as their truth values. An ordinary distribution when no outcome holds an outcome of a draw.

sample

A draw when sampling: an outcome chosen with its probability, with a value for each of its draws, from its distribution where it can be.