Comonotonicity

In probability theory, comonotonicity mainly refers to the perfect positive dependence between the components of a random vector, essentially saying that they can be represented as increasing functions of a single random variable. In two dimensions it is also possible to consider perfect negative dependence, which is called countermonotonicity.

Comonotonicity is also related to the comonotonic additivity of the Choquet integral.

The concept of comonotonicity has applications in financial risk management and actuarial science, see e.g. and . In particular, the sum of the components X₁ + X₂ + · · · + X_n is the riskiest if the joint probability distribution of the random vector (X₁, X₂, . . . , X_n) is comonotonic. Furthermore, the α-quantile of the sum equals the sum of the α-quantiles of its components, hence comonotonic random variables are quantile-additive. In practical risk management terms it means that there is minimal (or eventually no) variance reduction from diversification.

For extensions of comonotonicity, see and .

Definitions

Comonotonicity of subsets of Rⁿ

A subset S of Rⁿ is called comonotonic (sometimes also nondecreasing) if, for all (x₁, x₂, . . . , x_n) and (y₁, y₂, . . . , y_n) in S with x_i < y_i for some i ∈ {1, 2, . . . , n}, it follows that x_j ≤ y_j for all j ∈ {1, 2, . . . , n}.

This means that S is a totally ordered set.

Comonotonicity of probability measures on Rⁿ

Let μ be a probability measure on the n-dimensional Euclidean space Rⁿ and let F denote its multivariate cumulative distribution function, that is

$F(x_1,\dots ,x_n):=\mu (\{(y_1,\dots ,y_n)\in {\mathbb{ℝ}}^n\mid y_1\le x_1,\dots ,y_n\le x_n\}),(x_1,\dots ,x_n)\in {\mathbb{ℝ}}^n.$

Furthermore, let F₁, . . . , F_n denote the cumulative distribution functions of the n one-dimensional marginal distributions of μ, that means

$F_i(x):=\mu (\{(y_1,\dots ,y_n)\in {\mathbb{ℝ}}^n\mid y_i\le x\}),x\in \mathbb{ℝ}$

for every i ∈ {1, 2, . . . , n}. Then μ is called comonotonic, if

$F(x_1,\dots ,x_n)=\underset{i\in \{1,\dots ,n\}}{\min }{F}_i(x_i),(x_1,\dots ,x_n)\in {\mathbb{ℝ}}^n.$

Note that the probability measure μ is comonotonic if and only if its support S is comonotonic according to the above definition.

Comonotonicity of Rⁿ-valued random vectors

An Rⁿ-valued random vector X = (X₁, . . . , X_n) is called comonotonic, if its multivariate distribution (the pushforward measure) is comonotonic, this means

$\mathrm{Pr}(X_1\le x_1,\dots ,X_n\le x_n)=\underset{i\in \{1,\dots ,n\}}{\min }\mathrm{Pr}(X_i\le x_i),(x_1,\dots ,x_n)\in {\mathbb{ℝ}}^n.$

Properties

An Rⁿ-valued random vector X = (X₁, . . . , X_n) is comonotonic if and only if it can be represented as

$(X_1,\dots ,X_n){=}_{\text{d}}(F_{X_1}^{-1}(U),\dots ,F_{X_n}^{-1}(U)),$

where =_d stands for equality in distribution, on the right-hand side are the left-continuous generalized inverses of the cumulative distribution functions F_X1, . . . , F_Xn, and U is a uniformly distributed random variable on the unit interval. More generally, a random vector is comonotonic if and only if it agrees in distribution with a random vector where all components are non-decreasing functions (or all are non-increasing functions) of the same random variable.

Upper bounds

Upper Fréchet–Hoeffding bound for cumulative distribution functions

Let X = (X₁, . . . , X_n) be an Rⁿ-valued random vector. Then, for every i ∈ {1, 2, . . . , n},

$\mathrm{Pr}(X_1\le x_1,\dots ,X_n\le x_n)\le \mathrm{Pr}(X_i\le x_i),(x_1,\dots ,x_n)\in {\mathbb{ℝ}}^n,$

hence

$\mathrm{Pr}(X_1\le x_1,\dots ,X_n\le x_n)\le \underset{i\in \{1,\dots ,n\}}{\min }\mathrm{Pr}(X_i\le x_i),(x_1,\dots ,x_n)\in {\mathbb{ℝ}}^n,$

with equality everywhere if and only if (X₁, . . . , X_n) is comonotonic.

Upper bound for the covariance

Let (X, Y) be a bivariate random vector such that the expected values of X, Y and the product XY exist. Let (X^*, Y^*) be a comonotonic bivariate random vector with the same one-dimensional marginal distributions as (X, Y). Then it follows from Höffding's formula for the covariance and the upper Fréchet–Hoeffding bound that

$\text{Cov}(X,Y)\le \text{Cov}(X^{*},Y^{*})$

and, correspondingly,

$\mathrm{E}[XY]\le \mathrm{E}[X^{*}{Y}^{*}]$

with equality if and only if (X, Y) is comonotonic.

Note that this result generalizes the rearrangement inequality and Chebyshev's sum inequality.

See also

• Copula (probability theory)