Folded-t and half-t distributions¶

Nonnegative distributions obtained by taking the absolute value of a Student-t variate, with the half-t arising from a centered symmetric t distribution restricted or folded at zero.

Version: v2 · 2026-09-19 · History
Domain-specific #: 4568
Domain group: Formal Sciences
Origin domain: Experimental Design & Statistics
Subdomain: Folded Distributions → Experimental Design & Statistics

Core Idea¶

Folded-t distributions are the laws of absolute values of possibly noncentered t variables; half-t is the centered symmetric case. The many-to-one absolute-value transformation combines the t density at y and −y for positive y, retaining heavy tails while removing sign. The abstraction is therefore identified by a declared carrier, a transformation or constraint over that carrier, and an invariant that tells an analyst whether the named structure is genuinely present.

The load-bearing residual is not the broad topic of statistics. It is absolute-value transformation of heavy-tailed Student distributions. That residual remains recognizable when examples, notation, scale, or implementation change, but it disappears if the carrier is mistyped, the condition that support is nonnegative and the density and normalization follow from both preimages under a stated t location-scale parameterization fails, a neighboring object is substituted, or notation and topical resemblance replace the constitutive test.

How would you explain it like I'm…

No faithful explanation at this level. A child-level picture can only say 'ignore the minus sign,' which collapses into folding any number or bell curve. It loses that this is a probability law built from a heavy-tailed Student t whose density sums both mirror-image sides.

The Folded Sand Pile

Imagine a number line with zero in the middle, and a pile of sand showing how often each number comes up: most sand is near the middle, but some is spread surprisingly far out on both sides. That shape is like a t-distribution. Now fold the number line in half at zero, so the left side lands on top of the right side and the sand piles add together. The new pile only covers zero and positive numbers, but it still has those long, far-reaching edges. That folded pile is a folded-t distribution, and if the original pile was perfectly centered at zero, it's called half-t.

Absolute Value of Heavy-Tailed t

A folded-t distribution describes the absolute value of a random variable that follows a Student t distribution, possibly shifted away from zero. The t distribution is like the bell curve but with heavier tails, meaning extreme values are more likely. Taking absolute values removes the sign, so the new distribution lives only on zero and positive numbers. For any positive value y, its probability density is the t density at y plus the t density at -y, since both map to y. The half-t is the special case where the original t is centered at zero and symmetric, so the two halves simply stack. The heavy tails survive the folding; only the sign is lost. To work with it correctly, you must state the location and scale used in the t distribution.

The folded-t distribution is the law of |X| where X follows a Student t location-scale distribution, possibly noncentered; the half-t is the centered, symmetric special case. Absolute value is a many-to-one map, so for y > 0 the folded density is f_t(y) + f_t(-y), combining the contributions of both preimages. The resulting distribution has support on the nonnegative reals and retains the heavy, polynomially decaying tails of the t while discarding sign. In the half-t case, the symmetry of the t makes the two preimage contributions equal, so the density is twice the t density on the positive half-line. Correct use requires a stated parameterization: location, scale and degrees of freedom. Checking an instance means verifying nonnegative support and that density and normalization follow from both preimages under that parameterization. It should not be confused with neighboring objects such as the half-normal, which folds a light-tailed normal instead.

Scope of Application¶

Folded-t and half-t distributions belongs to statistics and is useful where the analyst can specify a Student-t random variable with degrees of freedom, location and scale, absolute-value map Y=|X|, nonnegative support, paired positive and negative preimages, folded density and centered half-t special case, then evaluate support is nonnegative and the density and normalization follow from both preimages under a stated t location-scale parameterization. The scope is broad within that domain but bounded by the need for support is nonnegative and the density and normalization follow from both preimages under a stated t location-scale parameterization. The entry records a descriptive analytical identity; practical use requires the governing domain's evidence, standards, and safety obligations.

Clarity¶

The abstraction clarifies a crowded vocabulary by making support is nonnegative and the density and normalization follow from both preimages under a stated t location-scale parameterization the center of the account. A claim should name the carrier, the governing operation or relation, the applicable assumptions, and the recognition test. A bare label is insufficient because the name Folded-t and half-t distributions can be used for a formal identity, an implementation, or a neighboring result unless carrier and convention are stated.

Manages Complexity¶

Without the abstraction, an analyst must reason directly over many local details: the carrier roles, admissibility assumptions, competing conventions, derived invariants, boundary cases, and proof or validation obligations specific to Folded-t and half-t distributions. Folded-t and half-t distributions compresses them into the roles in the structural signature. That compression permits comparison across instances without erasing the variables that determine validity. It also exposes which details may be varied safely and which are constitutive.

Abstract Reasoning¶

Identify the carrier. State what the elements, states, objects, or observations are: a Student-t random variable with degrees of freedom, location and scale, absolute-value map Y=|X|, nonnegative support, paired positive and negative preimages, folded density and centered half-t special case. Reject examples whose alleged carrier belongs to a different problem.
Lock the constitutive rule. Express support is nonnegative and the density and normalization follow from both preimages under a stated t location-scale parameterization independently of one notation or implementation.

Knowledge Transfer¶

Knowledge transfers strongly among subfields of statistics because they reuse a Student-t random variable with degrees of freedom, location and scale, absolute-value map Y=|X|, nonnegative support, paired positive and negative preimages, folded density and centered half-t special case, The many-to-one absolute-value transformation combines the t density at y and −y for positive y, retaining heavy tails while removing sign., and type the carrier, state every parameter and convention in the definition, test that support is nonnegative and the density and normalization follow from both preimages under a stated t location-scale parameterization, compare the nearest accepted identity, and report counterexamples, uncertainty, and limiting cases.

Relationships to Other Abstractions¶

Current abstraction Folded-t and half-t distributions Domain-specific

Parents (1) — outgoing typed relations

Folded-t and half-t distributions is a kind of Transformation Prime

The distributions arise by applying an absolute-value transformation to a t variable.

Hierarchy path (1) — routes to 1 parentless root

Folded-t and half-t distributions → is a kind of Transformation → is a kind of, typical Function (Mapping)

Neighborhood in Abstraction Space¶

Folded-t and half-t distributions sits in a crowded region of the domain-specific corpus (16^th percentile for distinctiveness): several abstractions share nearly its structure, so a description that fits it tends to fit its neighbors too.

Family — Probability Distributions & Second-Moment Structures (32 abstractions)

Nearest neighbors

Location parameter — 0.91
Exchangeable random variables — 0.91
Modified half-normal distribution — 0.91
Uncorrelatedness — 0.90
Correspondence analysis — 0.90

Computed from structural-signature embeddings · 2026-10-25