pith. sign in

arxiv: 2606.21042 · v1 · pith:UBYUANCNnew · submitted 2026-06-19 · 🧮 math.PR · q-fin.RM

Absolute Continuity of Monotone Aggregations under Positive Regression Dependence

Ben Goldys , Max Nendel This is my paper

Pith reviewed 2026-06-26 13:50 UTC · model grok-4.3

classification 🧮 math.PR q-fin.RM
keywords absolute continuitypositive regression dependencestochastic ordermonotone aggregationrisk measurespushforward measuresconditional distributionsdependent random variables
0
0 comments X

The pith

Under positive regression dependence, monotone maps g(X,Y) yield absolutely continuous distributions even without independence or joint densities.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper proves that absolute continuity of X combined with stochastic monotonicity of the conditional law of Y given X is enough to make g(X,Y) absolutely continuous whenever g is coordinatewise monotone in Y and satisfies a uniform lower-increment condition in X. This matters for applications such as risk aggregation because it removes any requirement that the variables be independent or possess a joint density while still guaranteeing a continuous distribution for the aggregate. The argument works by preventing atoms through the interaction of the increment condition and the ordering of the conditional distributions, and it extends immediately to Y taking values in any measurable space equipped with a reflexive relation.

Core claim

Suppose X has an absolutely continuous distribution and the conditional distribution of the R^d-valued random vector Y given X=x is nondecreasing in x in the usual stochastic order. Then any Borel map g from R times R^d to R that is coordinatewise nondecreasing in Y and satisfies a uniform lower-increment condition in X produces an absolutely continuous random variable g(X,Y). The result requires neither independence nor a joint density and allows the marginal law of Y to be arbitrary.

What carries the argument

The uniform lower-increment condition on g in the X variable, which forces positive increase when X increases and interacts with the stochastic monotonicity of the conditional distributions of Y to rule out atoms.

If this is right

  • Monotone risk aggregations remain absolutely continuous under positive regression dependence.
  • Regularization by convolution extends beyond the independent case to this class of dependent random vectors.
  • The absolute-continuity conclusion holds when the space of Y is replaced by any measurable space with a reflexive binary relation.
  • No joint density between X and Y is needed for the conclusion.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same monotonicity structure could be used to obtain explicit density bounds or modulus-of-continuity estimates for g(X,Y).
  • Risk-management calculations that previously required independence assumptions can be relaxed to positive regression dependence while retaining continuity of the aggregate.
  • Numerical approximation schemes for such aggregates may exploit the guaranteed absolute continuity to avoid discrete artifacts.

Load-bearing premise

The conditional distributions of Y given X=x must be nondecreasing in the usual stochastic order as x grows, and g must increase by at least a fixed positive amount whenever X increases by one unit.

What would settle it

An explicit pair of conditional distributions that are stochastically monotone together with a monotone g that violates the uniform lower-increment condition, such that the resulting g(X,Y) places positive mass at a single point.

read the original abstract

In this paper, we provide a sufficient condition for the absolute continuity of one-dimensional push-forwards of dependent random vectors. Suppose that $X$ has an absolutely continuous distribution and that the conditional distribution of an $\mathbb{R}^d$-valued random vector $Y$ given $X=x$ is nondecreasing in $x\in \mathbb{R}$ in the usual stochastic order. For Borel maps $g\colon \mathbb{R}\times\mathbb{R}^d\to\mathbb{R}$ satisfying a coordinatewise monotonicity condition in $Y$ and a uniform lower-increment condition in $X$, we prove that $g(X,Y)$ has an absolutely continuous distribution. The result requires neither independence nor a joint density, and allows the marginal law of $Y$ to be completely arbitrary. Moreover, the result remains valid if $\mathbb{R}^d$ is replaced by an arbitrary measurable space endowed with a reflexive binary relation. We discuss consequences for monotone risk aggregation and extensions of the familiar regularization by convolution beyond independent random variables.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

0 major / 2 minor

Summary. The paper proves a sufficient condition for absolute continuity of the one-dimensional pushforward measure induced by g(X,Y). Under the assumptions that X is absolutely continuous on R, the conditional law of the R^d-valued Y given X=x is nondecreasing in x with respect to the usual stochastic order, and g is Borel, coordinatewise nondecreasing in the Y argument and satisfies a uniform lower-increment condition in the X argument, the law of g(X,Y) is absolutely continuous. The result requires neither joint densities nor independence, permits arbitrary marginals for Y, and extends verbatim to an arbitrary measurable space equipped with a reflexive binary relation in place of R^d. Applications to monotone risk aggregation and convolution-type regularization for dependent variables are discussed.

Significance. The result supplies a clean, assumption-light criterion guaranteeing absolute continuity for monotone aggregations under positive regression dependence. This is useful in risk management and stochastic ordering theory, where dependence is the rule rather than the exception and joint densities are unavailable. The generalization to reflexive relations on general spaces and the explicit avoidance of independence are genuine strengths; the derivation appears to rest directly on the listed monotonicity and increment hypotheses without hidden circularity or parameter fitting.

minor comments (2)
  1. [Abstract] The precise statement of the uniform lower-increment condition on g (abstract, paragraph 2) would benefit from an explicit display equation or a short illustrative example immediately after its introduction.
  2. Notation for the usual stochastic order and the reflexive relation in the general-space extension should be introduced once in a dedicated preliminary section rather than only in the statement of the main theorem.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for their positive summary, significance assessment, and recommendation to accept the manuscript.

Circularity Check

0 steps flagged

No significant circularity identified

full rationale

The paper states a sufficient condition for absolute continuity of the pushforward measure of g(X,Y) under the hypotheses that X is absolutely continuous, the conditional law of Y given X is nondecreasing in the usual stochastic order, and g satisfies coordinatewise monotonicity in Y together with a uniform lower-increment condition in X. The abstract and claim present this as a direct theorem proved from the listed assumptions on stochastic order and monotonicity; no equations reduce a derived quantity to a fitted parameter by construction, no self-citation is invoked as the sole justification for a load-bearing uniqueness or ansatz step, and the argument is not a renaming of a known empirical pattern. The derivation chain is therefore self-contained within the stated probabilistic hypotheses.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

Relies on standard measure-theoretic and stochastic-order axioms from probability; no free parameters, invented entities, or ad-hoc assumptions listed in the abstract.

axioms (2)
  • standard math Usual stochastic order on probability measures is reflexive and compatible with the given binary relation on the codomain.
    Invoked to define the nondecreasing conditional distributions and the extension to general measurable spaces.
  • standard math Borel measurability of g and absolute continuity of the law of X with respect to Lebesgue measure.
    Required for the push-forward to be well-defined and for the target property of absolute continuity.

pith-pipeline@v0.9.1-grok · 5702 in / 1331 out tokens · 20649 ms · 2026-06-26T13:50:17.105871+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

16 extracted references

  1. [1]

    Ambrosio, N

    L. Ambrosio, N. Gigli, and G. Savaré.Gradient Flows in Metric Spaces and in the Space of Probability Measures. Basel: Birkhäuser, 2nd edition, 2008

    2008

  2. [2]

    Birkhoff.Lattice Theory, volume 25 ofColloq

    G. Birkhoff.Lattice Theory, volume 25 ofColloq. Publ., Am. Math. Soc.American Mathematical Society (AMS), Providence, RI, 1979

    1979

  3. [3]

    V. I. Bogachev.Measure theory. Vol. I, II. Springer-Verlag, Berlin, 2007

    2007

  4. [4]

    Cont and T

    R. Cont and T. Kokholm. A consistent pricing model for index options and volatility derivatives. Mathematical Finance, 23(2):248–274, 2013

    2013

  5. [5]

    Cont and L

    R. Cont and L. Wagalath. Fire sales forensics: Measuring endogenous risk.Mathematical Finance, 26(4):835–866, 2016

    2016

  6. [6]

    S. R. Das and R. Uppal. The effect of systemic risk on international portfolio choice.Journal of Finance, 59(6):2809–2834, 2004

    2004

  7. [7]

    B.A.DaveyandH.A.Priestley.Introduction to Lattices and Order.Cambridge: CambridgeUniversity Press, 2nd edition, 2002

    2002

  8. [8]

    De Vecchi and M

    C. De Vecchi and M. Scherer. Pricing insurance contracts with an existing portfolio as background risk.Insur. Math. Econ., 122:180–193, 2025

    2025

  9. [9]

    Kallenberg.Foundations of modern probability

    O. Kallenberg.Foundations of modern probability. In 2 volumes, volume 99 ofProbab. Theory Stoch. Model.Cham: Springer, 3rd revised and expanded edition, 2021

    2021

  10. [10]

    E. L. Lehmann. Some concepts of dependence.Ann. Math. Stat., 37:1137–1153, 1966

    1966

  11. [11]

    Müller and M

    A. Müller and M. Scarsini. Stochastic comparison of random vectors with a common copula.Math. Oper. Res., 26(4):723–740, 2001

    2001

  12. [12]

    Müller and M

    A. Müller and M. Scarsini. Archimedean copulae and positive dependence.J. Multivariate Anal., 93(2):434–445, 2005

    2005

  13. [13]

    Nourdin and T

    I. Nourdin and T. Simon. On the absolute continuity of Lévy processes with drift.Ann. Probab., 34(3):1035–1051, 2006

    2006

  14. [14]

    Shaked and J

    M. Shaked and J. G. Shanthikumar.Stochastic orders. Springer Ser. Stat. New York, NY: Springer, 2007

    2007

  15. [15]

    J. W. Tukey. A problem of Berkson, and minimum variance orderly estimators.Ann. Math. Stat., 29:588–592, 1958

    1958

  16. [16]

    Villani.Optimal Transport

    C. Villani.Optimal Transport. Old and New, volume 338 ofGrundlehren Math. Wiss.Berlin: Springer, 2009. School of Mathematics and Statistics, The University of Sydney Email address:beniamin.goldys@sydney.edu.au Department of Statistics and Actuarial Science, University of W aterloo Email address:mnendel@uwaterloo.ca

    2009