Cokurtosis¶

Measure fourth-order joint variation by taking standardized expectations of products containing four centered random-variable factors, retaining how extreme deviations co-occur beyond covariance and coskewness.

Version: v3 · 2026-09-19 · History
Domain-specific #: 1497
Domain group: Formal Sciences
Origin domain: Experimental Design & Statistics
Subdomains: Multivariate Statistics, Higher Order Comoments → Experimental Design & Statistics

Core Idea¶

For standardized variables, a cokurtosis component is \(K_{ijkl}=E[(X_i-\mu_i)(X_j-\mu_j)(X_k-\mu_k)(X_l-\mu_l)]/(\sigma_i\sigma_j\sigma_k\sigma_l)\). With two variables, the nontrivial multiplicity patterns are (K_{1112}), (K_{1122}), and (K_{1222}). Centering removes location, multiplication binds four deviations from the same joint draw, expectation aggregates them under the joint law, and scale normalization removes measurement units. Repeated indices encode asymmetric or symmetric tail comovement, while the full tensor exposes effects that covariance cannot retain.

How would you explain it like I'm…

No faithful explanation at this level. All three generators agree that the only child-level picture is 'two things are extreme together', which is the vague tail-togetherness reading the core idea rejects in favor of an indexed, centered, standardized fourth-order cross moment.

Four-Way Spread Multiplying

Take pairs of measurements, like the height and weight of the same kid. For each kid, work out how far each number is from its average, and divide by its usual spread so the units go away. Now multiply four of these distances together—for example three from height and one from weight, or two from each—and average over all the kids. That average is a Cokurtosis number. Which mix of height and weight you used matters, so there are several different Cokurtosis numbers for one pair.

Fourth-Order Joint Moment

Covariance measures how two variables vary together using products of two deviations from their means. Cokurtosis goes to fourth order: it averages products of four standardized deviations drawn from the same observation, with each factor coming from one of the variables. With two variables, the interesting versions are three-of-one-and-one-of-the-other (two of these, asymmetric) and two-and-two (symmetric). Because deviations are raised to high powers, cokurtosis is very sensitive to large deviations, and it captures dependence that covariance cannot see. It is not a vague 'extremes happen together' idea: you must state which variables fill which of the four slots, subtract the means first, and scale each factor by its own standard deviation. Some definitions also subtract the value a normal distribution would give, and that convention has to be named.

For random variables X_i with means μ_i and standard deviations σ_i, a cokurtosis component is K_ijkl = E[(X_i−μ_i)(X_j−μ_j)(X_k−μ_k)(X_l−μ_l)] / (σ_iσ_jσ_kσ_l). Centering removes location, multiplying four deviations from the same joint draw binds them together, expectation aggregates under the joint distribution, and dividing by the scales removes units. For two variables the nontrivial patterns are K_1112, K_1122 and K_1222; repeated indices encode asymmetric (3-1) versus symmetric (2-2) tail comovement, and the full fourth-order tensor retains structure covariance discards. It requires paired observations under one joint law, finite fourth moments and nonzero scales. One must distinguish the raw standardized moment from the related fourth cumulant or an excess form that subtracts the Gaussian contribution, and name which convention is used. Sample estimates are highly sensitive to extreme observations. Uses include describing higher-order dependence, expanding a portfolio's fourth moment, and testing for non-Gaussian structure.

Scope of Application¶

Cokurtosis applies when the analyst can specify random variables with finite fourth moments and nonzero standard deviations, or a centered random vector with a declared fourth-moment tensor convention and establish that four factors from one joint distribution are centered, their index pattern is preserved, the expectation exists, and the chosen standardized-moment or fourth-cumulant convention is stated before Gaussian baselines or excess values are interpreted. The entry fixes the fourth-order comoment family and its conventions; it does not treat cokurtosis as a complete tail-risk measure or imply causation.

Clarity¶

A clear claim names the carrier, governing rule, assumptions, and recognition test. This matters because the prefix co- does not specify whether a source means a central moment, standardized statistic, or cumulant, and kurtosis itself has raw and excess conventions. The disciplined statement is that the object counts as Cokurtosis exactly when four factors from one joint distribution are centered, their index pattern is preserved, the expectation exists, and the chosen standardized-moment or fourth-cumulant convention is stated before Gaussian baselines or excess values are interpreted

Manages Complexity¶

The abstraction compresses two-variable repeated-index statistics, full multivariate fourth-moment tensors, standardized and unstandardized forms, cumulant and excess forms, sample estimators, factor approximations, and conditional time-varying comoments into a stable carrier, rule, invariant, and failure boundary. It makes comparison tractable while retaining the variables that control validity.

Abstract Reasoning¶

Type the carrier. Establish random variables with finite fourth moments and nonzero standard deviations, or a centered random vector with a declared fourth-moment tensor convention and reject examples from a different problem.
Lock the rule. Express that four factors from one joint distribution are centered, their index pattern is preserved, the expectation exists, and the chosen standardized-moment or fourth-cumulant convention is stated before Gaussian baselines or excess values are interpreted independently of one notation or implementation.

Knowledge Transfer¶

Transfer within multivariate statistics is strong when new cases preserve the same carrier, mechanism, and diagnostic. The move from For standardized jointly Gaussian \(X,Y\) with correlation \(\rho\), Isserlis' identity gives \(K(X,X,Y,Y)=1+2\rho^2\) and \(K(X,X,X,Y)=K(X,Y,Y,Y)=3\rho\). to For a portfolio return \(R=w^{\mathsf T}X\), its fourth central moment expands as \(\sum_{i,j,k,l}w_iw_jw_kw_l\,E[x_ix_jx_kx_l]\), so the cokurtosis tensor contributes to portfolio tail-shape calculations. demonstrates that continuity.

Relationships to Other Abstractions¶

Current abstraction Cokurtosis Domain-specific

Parents (1) — outgoing typed relations

Cokurtosis is a kind of Expected Value Prime

Every cokurtosis component is literally an expectation of a specified centered four-factor random quantity; What makes it its own entry: the indexed fourth-order cross central moment or explicitly related cumulant, not a vague claim that two variables have extreme values together.

Hierarchy paths (2) — routes to 2 parentless roots

Cokurtosis → is a kind of Expected Value → is a kind of Aggregation → is a kind of Transformation → is a kind of, typical Function (Mapping)

Show alternative path (1)

Neighborhood in Abstraction Space¶

Cokurtosis sits in a moderately populated region (41^st percentile for distinctiveness): it has near-neighbors but no dense thicket of look-alikes.

Family — Multivariate & Heavy-Tailed Distributions (16 abstractions)

Nearest neighbors

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