Indecomposable distribution¶
An indecomposable probability distribution cannot be represented as the convolution of two non-degenerate probability distributions, making it irreducible with respect to addition of independent random variables.
Core Idea¶
An indecomposable probability distribution is one that cannot be expressed as the convolution of two nondegenerate probability distributions, equivalently as the law of X+Y for independent, nonconstant random variables X and Y. Decomposability concerns additive independent factors of the whole distribution, not whether a random variable can be written algebraically from dependent quantities. Degenerate point masses are excluded as trivial factors because any law could otherwise be convolved with a constant shift. Characteristic functions turn convolution into multiplication, so decomposing a distribution corresponds to factoring its characteristic function into characteristic functions of nondegenerate laws.
Scope of Application¶
-
Probability factorization. Candidate additive components are tested under a declared class of distributions.
-
Characteristic-function analysis. Convolution becomes multiplication, but every factor must itself be a valid characteristic function.
-
Finite-support classification. Support cardinality and geometry can obstruct nontrivial sums.
-
Discrete mass constraints. Probability values determine whether a proposed Bernoulli or lattice factorization is possible.
-
Additive building blocks. Indecomposable laws serve as atoms under independent summation within a chosen setting.
Clarity¶
Indecomposable distribution cannot be represented as the convolution of two nondegenerate probability laws, equivalently as a sum of independent nonconstant random variables. The independence and exclusion of trivial point-mass factors are essential; an algebraic or dependent decomposition of one random variable is irrelevant. Characteristic-function factorization makes the condition testable in another domain.
Manages Complexity¶
Indecomposable distribution compresses additive factorization to the question whether a probability law has any two nondegenerate convolution factors. Characteristic functions turn that question into multiplicative factorization, while support and atom structure provide quick obstructions. Decomposable, indecomposable, and infinitely divisible branches separate progressively different properties. The probabilist can treat indecomposable laws as building blocks without examining every representation of a random variable, because only sums of independent nonconstant factors count.
Abstract Reasoning¶
Factorization move. Ask whether a probability distribution can be expressed as the convolution of two nondegenerate distributions. Contradiction move. Assume such a decomposition and use characteristic functions, support, moments, zeros, or regularity to rule it out. Construction move. Build examples whose support or transform forbids nontrivial convolution factors. Comparison move. Distinguish indecomposable laws from infinitely divisible, stable, prime-like, or merely non-Gaussian distributions. Boundary move. Indecomposability concerns convolution structure, not whether a random variable has independent components in some representation, and degenerate point-mass factors are normally excluded.
Knowledge Transfer¶
Within the home domain. Indecomposable distributions transfer across probability theory, convolution factorization, limit laws, and characteristic-function analysis as nondegenerate laws that cannot be expressed as a convolution of two nondegenerate probability distributions. Support, transform, factor, convolution, and degeneracy retain exact roles. Beyond the home domain (C — formal class). The property applies literally to any probability law under the stated factorization convention. Its boundary is semantic: it does not mean a random variable lacks component representations, statistical independence is not the same question, and “indecomposable” in algebra or representation theory denotes different structures despite family resemblance.
Relationships to Other Abstractions¶
Current abstraction Indecomposable distribution Domain-specific
Parents (1) — more general patterns this builds on
-
Indecomposable distribution is a kind of Probability Distribution Domain-specific
Indecomposable distribution is a domain-specific kind of Probability Distribution: An indecomposable probability distribution cannot be represented as the convolution of two non-degenerate probability distributions, making it irreducible with respect to addition of independent random variables.
Hierarchy paths (5) — routes to 3 parentless roots
- Indecomposable distribution → Probability Distribution → Random Variable → Function (Mapping)
- Indecomposable distribution → Probability Distribution → Probability → Measure → Set and Membership
- Indecomposable distribution → Probability Distribution → Probability → Measure → Aggregation → Micro Macro Linkage
- Indecomposable distribution → Probability Distribution → Random Variable → Probability → Measure → Set and Membership
- Indecomposable distribution → Probability Distribution → Random Variable → Probability → Measure → Aggregation → Micro Macro Linkage
Neighborhood in Abstraction Space¶
Indecomposable distribution sits in a sparse region of the domain-specific corpus (82nd percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.
Family — Composite Probability Distributions (5 abstractions)
Nearest neighbors
- Kolmogorov's Three-Series Theorem — 0.84
- Variational Bayesian Methods — 0.82
- Lévy's continuity theorem — 0.82
- Exponentially Modified Gaussian Distribution — 0.81
- Delaporte Distribution — 0.81
Computed from structural-signature embeddings · 2026-10-08