pith. sign in

arxiv: 2606.29041 · v1 · pith:UEM3DODRnew · submitted 2026-06-27 · 📊 stat.ME · math.ST· stat.TH

On Modeling Cylindrical Data with a Discrete Circular Component and Its Environmental Applications

Brajesh Kumar Dhakad , Jayant Jha This is my paper

Pith reviewed 2026-06-30 08:28 UTC · model grok-4.3

classification 📊 stat.ME math.STstat.TH
keywords circular datadiscrete circular variablecylindrical datawrapped geometric distributionWeibull distributioncircular-linear regressionenvironmental statisticsjoint modeling
0
0 comments X

The pith

A model joins a wrapped symmetric geometric distribution for discrete circular data with a Weibull for the linear component via a trigonometric link, producing closed-form joint and conditional distributions plus explicit regression moments

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

The paper constructs a joint distribution for a discrete circular variable observed on equally spaced directions and a continuous linear variable. It uses the wrapped symmetric geometric distribution for the circular margin, the Weibull for the linear margin, and a trigonometric function to induce dependence. This choice delivers closed-form expressions for the joint, marginal, and conditional distributions, permits direct sampling by inverting the Weibull cumulative distribution function, and supplies explicit conditional moments that define a regression model. The construction is motivated by environmental data where directions are recorded only at discrete angles and linear measurements such as speed accompany them.

Core claim

The model formed by a wrapped symmetric geometric distribution for the discrete circular component, a Weibull distribution for the linear component, and a trigonometric linking function admits closed-form joint, marginal, and conditional distributions, allows direct inverse-transform sampling, and yields explicit expressions for conditional moments that support a flexible circular-linear regression framework.

What carries the argument

The trigonometric linking function that couples the wrapped symmetric geometric circular margin to the Weibull linear margin.

If this is right

  • Joint, marginal, and conditional distributions exist in closed form.
  • Samples from the joint distribution can be generated directly by the inverse transform method applied to the Weibull.
  • Explicit formulas for conditional moments enable a parametric circular-linear regression.
  • The conditional mean and variance are monotonic functions of the dependence parameters.

Where Pith is reading between the lines

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

  • The same linking construction could be paired with other continuous distributions that admit inverse-transform sampling if closed-form conditionals are not required.
  • The model supplies a natural baseline for comparing against nonparametric or mixture approaches when discrete circular data arise in non-environmental settings such as sensor readings or navigation.
  • Because parameters have direct interpretations in terms of marginal shape and dependence strength, the framework lends itself to sensitivity checks by varying the linking coefficient while holding margins fixed.

Load-bearing premise

The wrapped symmetric geometric distribution together with the chosen trigonometric linking function is flexible enough to capture the dependence structure present in the target environmental datasets.

What would settle it

Applying the fitted model to the two empirical environmental datasets and finding that the observed conditional moments or dependence patterns deviate substantially from the closed-form predictions would falsify the claim of adequate flexibility.

read the original abstract

Standard statistical methods are often inadequate for modeling the joint dependence between linear and circular variables, and existing methods for modeling this dependence are designed only for continuous variables. However, circular data are frequently observed on a finite set of equally spaced directions, either due to rounding prior to reporting or because of the experimental design employed for data collection. To address this gap, we propose a flexible, analytically tractable model for jointly representing a discrete circular and a continuous linear variable. The construction combines a wrapped symmetric geometric distribution, a Weibull distribution, and a trigonometric linking function. This formulation yields closed-form expressions for the joint, marginal, and conditional distributions. The choice of the Weibull distribution facilitates direct sample generation using the inverse transform technique. Additionally, it provides explicit expressions for conditional moments, enabling a flexible circular-linear regression framework. We detail the theoretical interpretation of the model parameters, mathematically establishing the monotonicity of the conditional mean and variance with respect to the dependence parameters. The performance of the estimators is demonstrated through extensive simulations, and the utility of the model is illustrated by analyzing two empirical environmental datasets.

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

2 major / 2 minor

Summary. The paper proposes a joint model for a discrete circular variable (using a wrapped symmetric geometric distribution) and a continuous linear variable (Weibull), connected via a trigonometric linking function. It claims this yields closed-form expressions for the joint, marginal, and conditional distributions, explicit conditional moments for regression, direct sampling via inverse transform, and mathematically established monotonicity of conditional mean and variance with respect to dependence parameters. The model is evaluated via simulations and applied to two environmental datasets.

Significance. If the derivations hold and the linking function proves sufficiently flexible, the work fills a noted gap in tractable modeling of mixed discrete-circular and continuous-linear data, which arises frequently in environmental applications due to rounding or design constraints. The closed-form results and explicit moments are a clear strength, enabling regression and inference without numerical integration, and the Weibull choice supports straightforward simulation.

major comments (2)
  1. [Model construction] Model construction (likely §2–3): The central claim that the model is useful for environmental applications rests on the trigonometric linking function being flexible enough to capture the dependence in the target datasets. However, trigonometric links typically induce symmetric, periodic dependence; the manuscript does not provide a direct test (e.g., residual analysis or comparison against asymmetric alternatives) showing that this form adequately represents the possibly asymmetric or higher-order associations present in the empirical data.
  2. [Theoretical interpretation and monotonicity] Monotonicity results (abstract and theoretical section): The paper states it mathematically establishes monotonicity of conditional mean and variance w.r.t. dependence parameters, but without the explicit derivation steps or verification that the trigonometric link preserves these properties under the wrapped geometric margin, it is difficult to confirm the result is not an artifact of the specific parametrization.
minor comments (2)
  1. [Abstract] The abstract refers to 'two empirical environmental datasets' without naming them or summarizing their key features (e.g., sample size, observed dependence patterns); this should be added for context.
  2. [Model section] Notation for the trigonometric link parameters (e.g., ρ, θ_circ) should be introduced with an explicit functional form in the model section to aid reproducibility.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the detailed and constructive report. We address each major comment below, indicating planned revisions where appropriate.

read point-by-point responses

  1. Referee: [Model construction] Model construction (likely §2–3): The central claim that the model is useful for environmental applications rests on the trigonometric linking function being flexible enough to capture the dependence in the target datasets. However, trigonometric links typically induce symmetric, periodic dependence; the manuscript does not provide a direct test (e.g., residual analysis or comparison against asymmetric alternatives) showing that this form adequately represents the possibly asymmetric or higher-order associations present in the empirical data.

    Authors: We agree that additional empirical checks would strengthen the application. In the revised manuscript we will add residual diagnostics for both datasets and a brief comparison against a non-trigonometric (asymmetric) alternative to confirm that the chosen link adequately captures the observed dependence structure. revision: yes

  2. Referee: [Theoretical interpretation and monotonicity] Monotonicity results (abstract and theoretical section): The paper states it mathematically establishes monotonicity of conditional mean and variance w.r.t. dependence parameters, but without the explicit derivation steps or verification that the trigonometric link preserves these properties under the wrapped geometric margin, it is difficult to confirm the result is not an artifact of the specific parametrization.

    Authors: The monotonicity follows from direct differentiation of the closed-form conditional moments; the sign of each derivative is determined by the monotonicity properties of the wrapped symmetric geometric pmf combined with the trigonometric link. We will expand the relevant theoretical section to display the full differentiation steps and the explicit verification for the wrapped geometric case. revision: yes

Circularity Check

0 steps flagged

No circularity; model built from standard distributions with derived closed forms

full rationale

The paper constructs the joint model explicitly from a wrapped symmetric geometric distribution for the discrete circular margin, a Weibull for the linear margin, and a trigonometric linking function. Closed-form joint/marginal/conditional distributions and conditional moments are stated to follow directly from this choice of components (abstract). No equations or text in the provided material reduce any claimed result to a fitted parameter, self-citation chain, or definitional equivalence. The derivation chain is self-contained against external benchmarks and does not invoke prior author work as a uniqueness theorem or ansatz source. This is the normal case of an honest new construction; score 0 is appropriate.

Axiom & Free-Parameter Ledger

3 free parameters · 2 axioms · 0 invented entities

The model rests on the standard properties of the wrapped symmetric geometric and Weibull distributions plus the assumption that a trigonometric link can be chosen to produce tractable conditionals. No new entities are postulated. Free parameters are the usual shape/scale parameters of the two distributions plus the dependence parameters of the link function; these are fitted rather than derived.

free parameters (3)
  • Weibull shape and scale
    Standard parameters of the linear marginal; must be estimated from data.
  • Wrapped geometric success probability
    Controls the discrete circular marginal; fitted to data.
  • Trigonometric link dependence parameters
    Control the strength and form of circular-linear dependence; introduced to link the two marginals.
axioms (2)
  • standard math The wrapped symmetric geometric distribution is a valid probability distribution on the discrete circle.
    Invoked when the circular component is defined.
  • standard math The Weibull distribution admits an inverse transform for sampling.
    Used to claim direct sample generation.

pith-pipeline@v0.9.1-grok · 5723 in / 1459 out tokens · 49935 ms · 2026-06-30T08:28:06.979789+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

15 extracted references · 2 canonical work pages

  1. [1]

    Econometrics and Statistics 4, 91–104 (2017)

    Abe, T., Ley, C.: A tractable, parsimonious and flexible model for cylindrical data, with applications. Econometrics and Statistics 4, 91–104 (2017)

    2017

  2. [2]

    Courier Dover Publications, New York (2015)

    Artin, E.: The Gamma Function. Courier Dover Publications, New York (2015)

    2015

  3. [3]

    https://doi.org/10.48550/arXiv.2602.12842

    Dhakad, B.K., Jha, J., Mukherjee, D.: Some bivariate distributions on a discrete torus with application to wind direction datasets (2026). https://doi.org/10.48550/arXiv.2602.12842

    work page doi:10.48550/arxiv.2602.12842 2026

  4. [4]

    Cambridge University Press, Cambridge, UK (1993)

    Fisher, N.I.: Statistical Analysis of Circular Data. Cambridge University Press, Cambridge, UK (1993)

    1993

  5. [5]

    Stochastic Environmental Research and Risk Assessment 27(5), 1055–1067 (2013)

    Garc´ ıa-Portugu´ es, E., Crujeiras, R.M., Gonz´ alez-Manteiga, W.: Exploring wind direction and so2 concentration by circular-linear density estimation. Stochastic Environmental Research and Risk Assessment 27(5), 1055–1067 (2013)

    2013

  6. [6]

    Environmetrics 29(2), 2490 (2018)

    Jha, J., Biswas, A.: Distribution and time-series modelling of ordinal circular data. Environmetrics 29(2), 2490 (2018)

    2018

  7. [7]

    Series on Multivariate Analysis, vol

    Jammalamadaka, S.R., SenGupta, A.: Topics in Circular Statistics. Series on Multivariate Analysis, vol. 5. WORLD SCIENTIFIC, Singapore (2001)

    2001

  8. [8]

    Journal of the American Statistical Association 73(363), 602–606 (1978)

    Johnson, R.A., Wehrly, T.E.: Some angular-linear distributions and related regression models. Journal of the American Statistical Association 73(363), 602–606 (1978)

    1978

  9. [9]

    Journal of Statistical Planning and Inference 138(11), 3538–3549 (2008) 22

    Kato, S., Shimizu, K.: Dependent models for observations which include angular ones. Journal of Statistical Planning and Inference 138(11), 3538–3549 (2008) 22

    2008

  10. [10]

    Journal of the Royal Statistical Society Series B: Statistical Methodology 37(3), 349–371 (1975)

    Mardia, K.V.: Statistics of directional data. Journal of the Royal Statistical Society Series B: Statistical Methodology 37(3), 349–371 (1975)

    1975

  11. [11]

    Biometrika, 403–405 (1976)

    Mardia, K.: Linear-circular correlation coefficients and rhythmometry. Biometrika, 403–405 (1976)

    1976

  12. [12]

    John Wiley & Sons, Chichester, UK (2009)

    Mardia, K.V., Jupp, P.E.: Directional Statistics. John Wiley & Sons, Chichester, UK (2009)

    2009

  13. [13]

    Jour- nal of the Royal Statistical Society Series B: Statistical Methodology 40(2), 229–233 (1978)

    Mardia, K., Sutton, T.: A model for cylindrical variables with applications. Jour- nal of the Royal Statistical Society Series B: Statistical Methodology 40(2), 229–233 (1978)

    1978

  14. [14]

    on the correlation of characters not quantitatively measurable

    Pearson, K.: Mathematical contributions to the theory of evolution.—VII. on the correlation of characters not quantitatively measurable. Philosoph- ical Transactions of the Royal Society of London. Series A, Contain- ing Papers of a Mathematical or Physical Character 195, 1–47 (1901) https://doi.org/10.1098/rsta.1900.0022

    work page doi:10.1098/rsta.1900.0022 1901

  15. [15]

    (eds.): Directional Statistics for Innovative Applica- tions

    SenGupta, A., Arnold, B.C. (eds.): Directional Statistics for Innovative Applica- tions. Forum for Interdisciplinary Mathematics. Springer, Singapore (2022) 23

    2022