pith. sign in

arxiv: 2607.01558 · v1 · pith:MSAP5UHSnew · submitted 2026-07-02 · 📊 stat.ME · math.ST· stat.AP· stat.TH

Lancaster copulas

Angelo Efoevi Koudou , Yves I. Ngounou Bakam , Denys Pommeret This is my paper

Pith reviewed 2026-07-03 00:42 UTC · model grok-4.3

classification 📊 stat.ME math.STstat.APstat.TH
keywords Lancaster copulasorthogonal expansionscopula densityseries representationstruncation effectsdependence modelingcontinuous Lancaster probabilities
0
0 comments X

The pith

Lancaster copulas are built from orthogonal expansions of continuous Lancaster probabilities, yielding series representations for the copula and density that remain accurate under low-order truncation.

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

The paper defines a new family of copulas by assembling orthogonal expansions of continuous Lancaster probabilities into dependence functions. It supplies explicit infinite-series formulas for both the copula and the associated density. The authors then analyze the consequences of truncating those series and test the resulting approximations on numerical examples, finding that low-order cuts already reproduce the target dependence closely. A reader would care because the construction supplies a direct, expandable route to new copula families whose computational cost can be controlled by choosing how many terms to keep.

Core claim

We introduce a new copula class, called Lancaster copulas, built from orthogonal expansions of continuous Lancaster probabilities. We derive infinite-series representations for the copula and its density, study truncation effects, and show in numerical experiments that low-order truncations already provide accurate approximation.

What carries the argument

Lancaster copulas assembled from orthogonal expansions of continuous Lancaster probabilities, which generate the series forms for the copula and density.

If this is right

  • The copula and its density each possess an explicit infinite-series representation.
  • Truncation of the series produces well-defined approximations whose accuracy can be examined term by term.
  • Numerical tests confirm that retaining only the lowest-order terms already yields close agreement with the target dependence.
  • The resulting family supplies a systematic method for generating copulas whose complexity is adjustable through the truncation order.

Where Pith is reading between the lines

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

  • The same expansion technique might be applied to other families of probabilities that admit orthogonal bases, producing further copula classes.
  • Explicit truncation-error bounds, if derived, would turn the numerical observations into a practical design rule for choosing series length.
  • Lancaster copulas may recover familiar parametric copulas as special cases when the underlying Lancaster probability is chosen appropriately.

Load-bearing premise

Orthogonal expansions of continuous Lancaster probabilities can be combined into functions that meet every requirement for a copula, including uniform marginal distributions on the unit interval.

What would settle it

A concrete Lancaster probability whose orthogonal expansion produces a function whose first marginal is not uniform on [0,1].

Figures

Figures reproduced from arXiv: 2607.01558 by Angelo Efoevi Koudou, Denys Pommeret, Yves I. Ngounou Bakam.

Figure 1
Figure 1. Figure 1: Density graphs and contour lines of DBVE [PITH_FULL_IMAGE:figures/full_fig_p012_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Density graphs and contour lines of Gaussian Lancaster copula [PITH_FULL_IMAGE:figures/full_fig_p013_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Density graphs and contour lines of gamma Lancaster copula den [PITH_FULL_IMAGE:figures/full_fig_p015_3.png] view at source ↗
read the original abstract

We introduce a new copula class, called Lancaster copulas, built from orthogonal expansions of continuous Lancaster probabilities. We derive infinite-series representations for the copula and its density, study truncation effects, and show in numerical experiments that low-order truncations already provide accurate approximation.

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 / 3 minor

Summary. The paper introduces Lancaster copulas constructed from orthogonal expansions of continuous Lancaster probabilities. It derives infinite-series representations for the copula C and its density c, analyzes truncation effects on these series, and presents numerical experiments showing that low-order truncations yield accurate approximations to the target dependence structures.

Significance. If the construction is valid, the work supplies a new parametric family of copulas with explicit series forms that facilitate both theoretical analysis and practical approximation. The truncation study and numerical validation are direct strengths, as they address usability of the infinite-series objects. This could be of interest in dependence modeling where flexible, series-based representations are needed.

minor comments (3)
  1. [§2-3] Clarify in §2 or §3 whether the orthogonal expansion is taken with respect to a specific weight function or measure, and state the precise conditions on the Lancaster probabilities that guarantee the resulting series defines a valid copula (uniform margins and 2-increasing property).
  2. [Numerical experiments section] In the numerical experiments, report the specific copula families or dependence parameters used as targets, and include quantitative error measures (e.g., sup-norm or integrated squared error) rather than qualitative statements of accuracy.
  3. [Discussion or conclusion] Add a short discussion of computational cost for evaluating the truncated series versus standard copula families, to help readers assess practical utility.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for their positive summary of our manuscript, recognition of the potential utility of the Lancaster copula construction, and recommendation of minor revision. We are pleased that the truncation analysis and numerical experiments were viewed as strengths.

Circularity Check

0 steps flagged

No significant circularity detected

full rationale

The paper constructs Lancaster copulas from orthogonal expansions of continuous Lancaster probabilities, then derives explicit infinite-series forms for the copula and density, analyzes truncation, and validates approximations numerically. No step reduces a claimed result to a fitted input renamed as prediction, a self-definitional loop, or a load-bearing self-citation chain; the copula axioms are addressed by the internal series derivations and experiments rather than assumed or imported. The derivation chain remains self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 1 invented entities

Abstract-only; ledger populated from stated construction steps only.

axioms (1)
  • domain assumption Continuous Lancaster probabilities admit orthogonal expansions that can be reassembled into valid copula functions.
    Central to the construction described in the abstract.
invented entities (1)
  • Lancaster copulas no independent evidence
    purpose: New class of copulas obtained from the orthogonal expansions.
    Introduced as the main object of study.

pith-pipeline@v0.9.1-grok · 5566 in / 1133 out tokens · 21172 ms · 2026-07-03T00:42:43.781662+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

53 extracted references

  1. [1]

    Bowman, Adrian W. (1984). An alternative method of cross-validation for the smoothing of density estimates.Biometrika, 71(2), 353–360

    1984

  2. [2]

    Taylor, Charles C. (1990). Orthogonal series estimators and cross- validation.Journal of Statistical Computation and Simulation, 37 (3-4), 151–158

    1990

  3. [3]

    Aas, K., Czado, C., Frigessi, A., & Bakken, H. (2009). Pair-copula con- structions of multiple dependence.Insurance: Mathematics and Eco- nomics, 44(2), 182–198

    2009

  4. [4]

    Ansari, J., & Rockel, M. (2024). Dependence properties of bivariate copula families.Dependence Modeling, 12(1), 20240002

    2024

  5. [5]

    Barrett, J., Lampard, D. (1955). An expansion for some second-order probability distributions and its application to noise problems.IRE Transactions on Information Theory, 1(1), 10–15

    1955

  6. [6]

    Brown, J.L. (1958). A criterion for the diagonal expansion of a second- order probability distribution in orthogonal polynomials.IRE Transac- tions on Information Theory, 4(4), 172–172

    1958

  7. [7]

    Buja, A. (1990). Remarks on functional canonical variates, alternating least squares methods and ACE.The Annals of Statistics, 19(3), 1032– 1069. 27

    1990

  8. [8]

    Bussgang, J. (1952). Crosscorrelation functions of amplitude-distorted gaussian signals.RLE Technical Reports

    1952

  9. [9]

    Cuadras, C.M. (2005). Continuous canonical correlation analysis.Re- search Letters in the Information and Mathematical Sciences8, 97-103

    2005

  10. [10]

    Dauxois, J., & Pousse, A. (1975). Une extension de l’analyse canonique. Quelques applications.Annales de l’IHP Probabilités et statistiques, 11, 355–379

    1975

  11. [11]

    Diaconis, P., & Griffiths, R. (2012). Exchangeable pairs of Bernoulli random variables, Krawtchouck polynomials, and Ehrenfest urns.Aus- tralian & New Zealand Journal of Statistics, 54(1), 81–101

    2012

  12. [12]

    Diaconis, P., Khare, K., & Saloff-Coste, L. (2008). Gibbs sampling, ex- ponentialfamiliesandorthogonalpolynomials.Statistical Science, 23(2), 151–178

    2008

  13. [13]

    Downton, F. (1970). Bivariate exponential distributions in reliability theory.Journal of the Royal Statistical Society: Series B (Methodologi- cal), 32(3), 408–417

    1970

  14. [14]

    (2002).Méthodes bayésiennes en statistique

    Droesbeke, J.J., Fine, J., & Saporta, G. (2002).Méthodes bayésiennes en statistique. Éditions Technip

    2002

  15. [15]

    Dueck, J., Edelmann, D., Richards, D. (2017). Distance correlation co- efficients for Lancaster distributions.Journal of Multivariate Analysis, 154, 19–39

    2017

  16. [16]

    Frahm, G., Junker, M., & Szimayer, A. (2003). Elliptical copulas: appli- cability and limitations.Statistics & Probability Letters, 63(3), 275–286

    2003

  17. [17]

    Goffard, P.O., Loisel, S., & Pommeret, D. (2017). Polynomial approxi- mations for bivariate aggregate claims amount probability distributions. Methodology and Computing in Applied Probability, 19, 151–174

    2017

  18. [18]

    Griffiths, B. (2009). Stochastic processes with orthogonal polynomial eigenfunctions.Journal of Computational and Applied Mathematics, 233(3), 739–744

    2009

  19. [19]

    Griffiths, R.C., & Spanò, D. (2013). Orthogonal polynomial kernels and canonical correlations for Dirichlet measures.Bernoulli, 19(2), 548–598

    2013

  20. [20]

    Gudendorf, G., & Segers, J. (2010). Extreme-value copulas. In:Copula Theory and Its Applications. Springer 127–145. 28

    2010

  21. [21]

    Guzmics, S., & Pflug, G.C. (2020). A new extreme value copula and new families of univariate distributions based on Freund’s exponential model.Dependence Modeling, 8, 330–360

    2020

  22. [22]

    He, F., Yarahmadi, A., & Soleymani, F. (2024). Investigation of multi- variate pairs trading under copula approach with mixture distribution. Applied Mathematics and Computation, 472, 128635

    2024

  23. [23]

    Hernández-Maldonado, V.M., Erdely, A., Díaz-Viera, M., & Rios, L. (2024). Fast procedure to compute empirical and Bernstein copulas.Ap- plied Mathematics and Computation, 477, 128827

    2024

  24. [24]

    Hofert, M., Mächler, M., & McNeil, A.J. (2013). Archimedean copulas in high dimensions: Estimators and numerical challenges motivated by financial applications.Journal de la Société Française de Statistique, 154(1), 25–63

    2013

  25. [25]

    (2014).Dependence Modeling with Copulas

    Joe„ J. (2014).Dependence Modeling with Copulas. CRC Press

    2014

  26. [26]

    (2004).Continuous Multi- variate Distributions, Volume 1: Models and Applications

    Kotz, S., Balakrishnan, N., & Johnson, N.L. (2004).Continuous Multi- variate Distributions, Volume 1: Models and Applications. John Wiley & Sons

    2004

  27. [27]

    Koudou, A.E., & Pommeret, D. (2000). A construction of Lancaster probabilities with margins in the multidimensional Meixner class.Aus- tralian & New Zealand Journal of Statistics, 42(1), 59–66

    2000

  28. [28]

    (1995).Problèmes de marges et familles exponentielles naturelles

    Koudou., A.E. (1995).Problèmes de marges et familles exponentielles naturelles. PhD thesis, Toulouse, 1995

    1995

  29. [29]

    Probabilités de Lancaster.Expositiones Mathe- maticae, 14, 247–276

    Koudou, A.E., (1996). Probabilités de Lancaster.Expositiones Mathe- maticae, 14, 247–276

    1996

  30. [30]

    Lancaster, H. (1958). The structure of bivariate distributions.The An- nals of Mathematical Statistics, 29(3), 719–736

    1958

  31. [31]

    Lancaster, H.O. (1963). Correlations and canonical forms of bivariate distributions.The Annals of Mathematical Statistics, 34(2), 532–538

    1963

  32. [32]

    Lancaster, H.O. (1975). Joint probability distributions in the Meixner classes.Journal of the Royal Statistical Society: Series B (Methodologi- cal), 37(3), 434–443

    1975

  33. [33]

    Longla, M.(2024).Newcopulafamiliesandmixingproperties.Statistical Papers65, 4331–4363. 29

    2024

  34. [34]

    & Palma, F

    Mena, R.H. & Palma, F. (2020). Continuous-time Markov processes, orthogonal polynomials and Lancaster probabilities.ESAIM: Probability and Statistics, 24(1), 100–112

    2020

  35. [35]

    Morris, C.N. (1982). Natural exponential families with quadratic vari- ance functions.Annals of Statistics, 10, 65–80

    1982

  36. [36]

    Muia, M.N., Longla, M. (2025). A Point on Discrete versus Continuous State-Space Markov Chains.Dependence Modelling, 13, 1-23

    2025

  37. [37]

    (2007).An Introduction to Copulas

    Nelsen., R.B. (2007).An Introduction to Copulas. Springer

    2007

  38. [38]

    Ngounou Bakam, Y.I., & Pommeret, D. (2025). Nonparametric estima- tion of copulas and copula densities by orthogonal projections.Econo- metrics and Statistics, 36, 90–118

    2025

  39. [39]

    Pfeifer, D., Mandle, A., Ragulina, O., & Girschig, C. (2019). New cop- ulas based on general partitions-of-unity (part III) — the continuous case.Dependence Modeling, 7(1), 181-201

    2019

  40. [40]

    Pommeret, D. (2004). A characterization of Lancaster probabilities with margins in a multivariate additive class.Sankhy¯ a: The Indian Journal of Statistics, 66(1), 1–19

    2004

  41. [41]

    Pommeret, D. (2005). Approximate polynomial expansion for joint den- sity.Applicationes Mathematicae, 32, 57–67

    2005

  42. [42]

    Quessy, J.F. (2024). General Construction of Multivariate Dependence StructureswithNonmonotoneMappingsandItsApplications.Statistical Science, 39(3), 391–408

    2024

  43. [43]

    Ren, H., Li, Q., Wu, Q., Zhang, C., Dou, Z., & Chen, J. (2022). Joint forecasting of multi-energy loads for a university based on copula theory and improved LSTM network.Energy Reports, 8, 605–612

    2022

  44. [44]

    Empirical choice of histograms and kernel density estimatorsScandinavian Journal of Statistics, 65–78

    Rudemo, Mats (1982). Empirical choice of histograms and kernel density estimatorsScandinavian Journal of Statistics, 65–78

    1982

  45. [45]

    Saminger-Platz, S., Kolesárová, A., Seliga, A., Mesiar, R., & Klement, E.P.(2024).Parameterizedtransformationsandtruncation: Whenisthe result a copula?.Journal of Computational and Applied Mathematics, 436, 115340. 30

    2024

  46. [46]

    Sarmanov, I. (1970). The approximate computation of the coefficient of correlation between functions of dependent random variables.Math. Notes Acad. Sciences USSR, 7, 373–377

    1970

  47. [47]

    Sarmanov, I. (1970). A gamma-correlation process and its properties. Doklady Akademii Nauk, 191, 30–32

    1970

  48. [48]

    Sarmanov, O., & Bratoeva, Z. (1967). Probabilistic properties of bi- linear expansions of Hermite polynomials.Theory of Probability & Its Applications, 12(3), 470–481

    1967

  49. [49]

    Schweitzer, B., Wolff, E.F. (1981). On nonparametric measures of de- pendence for random variables.Annals of Statistics, 9, 879-885

    1981

  50. [50]

    Sklar, M. (1959). Fonction de répartition à n dimensions et leurs marges. Annales de l’ISUP, 8, 229–231

    1959

  51. [51]

    Szego, G. (1975). Orthogonal Polynomials. 4th ed. (reprint). Ameri- can Mathematical Society Colloquium Publications, Vol. 23. American Mathematical Society, Providence, RI

    1975

  52. [52]

    Wong, E. (1964). The construction of a class of stationary Markoff pro- cesses.Stochastic Processes in Mathematical Physics and Engineering, 17, 264–276

    1964

  53. [53]

    Thomas, J

    Wong, E., & J. Thomas, J. (1962). On polynomial expansions of second- order distributions.Journal of the Society for Industrial and Applied Mathematics, 10(3), 507–516. 31

    1962