arXiv subjects

Samuel Soubeyrand

Publications and source records attributed to Samuel Soubeyrand.

11 recordsLinked to original sources

arXiv2026

Predicting disease severity and large-scale spread from coupled severity measurements and imperfect indicators: Application to beet yellows

Whether in human, animal, or plant health, effective disease management requires the ability to characterize disease dynamics across space and time. In this context, integrating indirect indicators with broad spatio-temporal coverage, even when they are noisy, can provide valuable complementary information to direct measurements, which are often sparse because they are more costly or intrusive to collect. In this article, we propose a statistical framework to leverage such indirect indicators to predict disease severity at the individual or local-scale level and reconstruct large-scale disease dynamics. This two-step approach is able to account for the specific characteristics of disease severity observations, including zero inflation and spatio-temporal structure. The first step relies on a stacked hurdle model based on multiple random forests to locally predict disease severity from the available indirect indicators. In the second step a semi-parametric spatio-temporal model is used to reconstruct large-scale epidemiological dynamics over space and time from the indicators-based predictions. The proposed methodology is designed to be both generic and modular, and is illustrated by a case study in plant health. This case study focuses on the monitoring of sugar beet yellows disease in France between 2019 and 2023 by combining sparse field measurements and satellite-based remote sensing data.

stat.ME↗

arXiv2026

A Top-Down Scale Approach for Multiscale Geographically and Temporally Weighted Regression

This paper proposes tds mgtwr, a multiscale geographically and temporally weighted regression (MGTWR) model with covariate-specific spatial and temporal scales. The approach combines a separable spatio-temporal kernel with a Top-Down Scale (TDS) calibration scheme, where spatial and temporal bandwidths are selected for each covariate through a coordinate-wise search over ordered grids guided by the corrected Akaike Information Criterion (AICc). By avoiding unconstrained multidimensional optimization, this strategy extends to the spatio-temporal setting the stabilizing properties of TDS calibration scheme Geniaux (2026). The multiscale backfitting procedure combines the Top-Down Scale calibration scheme with an adaptive, importance-driven update schedule that prioritizes covariates according to their current scale-normalized contribution to the fitted signal, thereby limiting the number of local recalibrations required and accelerating convergence while maintaining estimator fidelity. We also introduce a generic prediction method for MGWR and MGTWR based on kernel sharpening. Monte Carlo experiments show that modeling both space and time improves coefficient recovery and predictive accuracy relative to purely spatial multiscale models when temporal variation is present and sufficiently supported by the data. Gains increase with sample size and signal-to-noise ratio. Two empirical applications illustrate the method under contrasting regimes. For Beet Yellows severity, a plant epidemiology and pest management problem, multiscale spatial modeling is essential, while spatio-temporal extensions yield additional gains when temporal information is rich. In modeling house prices, MGTWR consistently outperforms spatial local and STVC models. In both cases, predictive performance rivals flexible machine-learning benchmarks while preserving interpretable spatio-temporal scales.

stat.ME↗

arXiv2025

Functional Laplace Transform of a Multivariate Hawkes Process, Subsequent Characteristics, and Numerical Approximations

Numerous studies grounded on Hawkes processes have been carried out in many fields including finance, biology and social network. Hawkes processes form a class of selfexciting simple point processes. In this article, we consider a general class of multivariate Hawkes processes envisioned to model dynamics of spatio-temporal epidemics. For this class, the igniting baseline intensity is time dependent and the exciting matrix function is a general one, making the model non-Markovian in most of the cases. In this article, we first provide the closed-form expression of the multivariate multi-temporal characteristic function of these Hawkes processes, extending in a natural way the classical single-time formula found in the Hawkes literature. Then, we use the infinitely divisible property of the Hawkes process to derive the equation system related to the probability distribution of counts at each single time, adapted to the general formulation of the Hawkes model considered in this article. Next, we provide closed-form formulas for the temporal structure of the two first moments of the process, which allows us to deduce an original expression of the multivariate covariance function at two distinct times, thereby extending existing results established for more restricted classes of Hawkes processes. Based on this expression, we analytically decompose the covariance at two distinct times into singular and continuous parts. We finish with brief numerical elements: We present a simple scheme for numerical approximations of the Laplace transform and the first two moments, and give examples of solutions of the different related integral equations. We also provides illustrative simulations of the multivariate Hawkes process for different model specifications.

math.PR↗

arXiv2024

Group Dispersal Modelling Revisited

In this paper we revisit the notion of grouped dispersal that have been introduced by Soubeyrand and co-authors \cite{soubeyrand2011patchy} to model the simultaneous (and hence dependent) dispersal of several propagules from a single source in a homogeneous environment. We built a time continuous measure valued process that takes into account the main feature of a grouped dispersal and derive its infinitesimal generator. To cope with the mutligeneration aspect associated to the demography we introduce two types of propagules in the description of the population which is one of the main innovations here. We also provide a rigorous description of the process and its generator. We derive as well, some large population asymptotics of the process unveilling the degenerate ultra parabolic system of PDE satisfied by the density of population. Finally, we also show that such a PDE system has a non-trivial solution which is unique in a certain functional space.

math.AP↗

arXiv2021

Spatial statistics and stochastic partial differential equations: a mechanistic viewpoint

The Stochastic Partial Differential Equation (SPDE) approach, now commonly used in spatial statistics to construct Gaussian random fields, is revisited from a mechanistic perspective based on the movement of microscopic particles, thereby relating pseudo-differential operators to dispersal kernels. We first establish a connection between Lévy flights and PDEs involving the Fractional Laplacian (FL) operator. The corresponding Fokker-Planck PDEs will serve as a basis to propose new generalisations by considering a general form of SPDE with terms accounting for dispersal, drift and reaction. We detail the difference between the FL operator (with or without linear reaction term) associated with a fat-tailed dispersal kernel and therefore describing long-distance dependencies, and the damped FL operator associated with a thin-tailed kernel, thus corresponding to short-distance dependencies. Then, SPDE-based random fields with non-stationary external spatially and temporally varying force are illustrated and nonlinear bistable reaction term are introduced. The physical meaning of the latter and possible applications are discussed. Returning to the particulate interpretation of the above-mentioned equations, we describe in a relatively simple case their links with point processes. We unravel the nature of the point processes they generate and show how such mechanistic models, associated to a probabilistic observation model, can be used in a hierarchical setting to estimate the parameters of the particle dynamics.

stat.ME↗

arXiv2020

Spatiotemporal large-scale networks shaped by air mass movements

The movement of atmospheric air masses can be seen as a continuous and complex flow of particles hovering over our planet. It can however be locally simplified by considering three-dimensional trajectories of air masses connecting distant areas of the globe during a given period of time. In this paper, we present a mathematical framework to construct spatial and spatiotemporal networks where the nodes are the subsets of a partition of a geographical area and the links between these nodes are inferred from sampled trajectories of air masses passing over and across the nodes. We propose different estimators of link intensities relying on different bio-physical hypotheses and covering adjustable time periods. This approach leads to a new class of spatiotemporal networks characterized by adjacency matrices. We applied the approach in two real geographical contexts: the watersheds of the French region Provence-Alpes-Côte d'Azur and the coastline of the Mediterranean Sea. The analysis of the constructed networks allowed identifying a marked seasonal pattern in air mass movements in the study areas. These constructed networks can be used to investigate issues, e.g., in aerobiology and epidemiology of airborne plant pathogens. Similar networks could be estimated from other types of trajectories, such as animal trajectories.

stat.AP↗

arXiv2020

A parsimonious model for spatial transmission and heterogeneity in the COVID-19 propagation

Raw data on the cumulative number of deaths at a country level generally indicate a spatially variable distribution of the incidence of COVID-19 disease. An important issue is to determine whether this spatial pattern is a consequence of environmental heterogeneities, such as the climatic conditions, during the course of the outbreak. Another fundamental issue is to understand the spatial spreading of COVID-19. To address these questions, we consider four candidate epidemiological models with varying complexity in terms of initial conditions, contact rates and non-local transmissions, and we fit them to French mortality data with a mixed probabilistic-ODE approach. Using standard statistical criteria, we select the model with non-local transmission corresponding to a diffusion on the graph of counties that depends on the geographic proximity, with time-dependent contact rate and spatially constant parameters. This original spatially parsimonious model suggests that in a geographically middle size centralized country such as France, once the epidemic is established, the effect of global processes such as restriction policies, sanitary measures and social distancing overwhelms the effect of local factors. Additionally, this modeling approach reveals the latent epidemiological dynamics including the local level of immunity, and allows us to evaluate the role of non-local interactions on the future spread of the disease. In view of its theoretical and numerical simplicity and its ability to accurately track the COVID-19 epidemic curves, the framework we develop here, in particular the non-local model and the associated estimation procedure, is of general interest in studying spatial dynamics of epidemics.

q-bio.PE↗

arXiv2020

Using early data to estimate the actual infection fatality ratio from COVID-19 in France (Running title: Infection fatality ratio from COVID-19)

The first cases of COVID-19 in France were detected on January 24, 2020. The number of screening tests carried out and the methodology used to target the patients tested do not allow for a direct computation of the real number of cases and the mortality rate.In this report, we develop a 'mechanistic-statistical' approach coupling a SIR ODE model describing the unobserved epidemiological dynamics, a probabilistic model describing the data acquisition process and a statistical inference method. The objective of this model is not to make forecasts but to estimate the real number of people infected with COVID-19 during the observation window in France and to deduce the mortality rate associated with the epidemic.Main results. The actual number of infected cases in France is probably much higher than the observations: we find here a factor x 15 (95%-CI: 4-33), which leads to a 5.2/1000 mortality rate (95%-CI: 1.5 / 1000-11.7/ 1000) at the end of the observation period. We find a R0 of 4.8, a high value which may be linked to the long viral shedding period of 20 days.

q-bio.PE↗

arXiv2020

Equilibrium and sensitivity analysis of a spatio-temporal host-vector epidemic model

Insect-borne diseases are diseases carried by insects affecting humans, animals or plants. They have the potential to generate massive outbreaks such as the Zika epidemic in 2015-2016 mostly distributed in the Americas, the Pacific and Southeast Asia, and the multi-foci outbreak caused by the bacterium {\it Xylella fastidiosa} in Europe in the 2010s. In this article, we propose and analyze the behavior of a spatially-explicit compartmental model adapted to pathosystems with fixed hosts and mobile vectors disseminating the disease. The behavior of this model based on a system of partial differential equations is complementarily characterized via a theoretical study of its equilibrium states and a numerical study of its transitive phase using global sensitivity analysis. The results are discussed in terms of implications concerning the surveillance and control of the disease over a medium-to-long temporal horizon.

math.AP↗

arXiv2019

Quick inference for log Gaussian Cox processes with non-stationary underlying random fields

For point patterns observed in natura, spatial heterogeneity is more the rule than the exception. In numerous applications, this can be mathematically handled by the flexible class of log Gaussian Cox processes (LGCPs); in brief, a LGCP is a Cox process driven by an underlying log Gaussian random field (log GRF). This allows the representation of point aggregation, point vacuum and intermediate situations, with more or less rapid transitions between these different states depending on the properties of GRF. Very often, the covariance function of the GRF is assumed to be stationary. In this article, we give two examples where the sizes (that is, the number of points) and the spatial extents of point clusters are allowed to vary in space. To tackle such features, we propose parametric and semiparametric models of non-stationary LGCPs where the non-stationarity is included in both the mean function and the covariance function of the GRF. Thus, in contrast to most other work on inhomogeneous LGCPs, second-order intensity-reweighted stationarity is not satisfied and the usual two step procedure for parameter estimation based on e.g. composite likelihood does not easily apply. Instead we propose a fast three step procedure based on composite likelihood. We apply our modelling and estimation framework to analyse datasets dealing with fish aggregation in a reservoir and with dispersal of biological particles.

math.ST↗

arXiv2018

Dating and localizing an invasion from post-introduction data and a coupled reaction-diffusion-absorption model

Invasion of new territories by alien organisms is of primary concern for environmental and health agencies and has been a core topic in mathematical modeling, in particular in the intents of reconstructing the past dynamics of the alien organisms and predicting their future spatial extents. Partial differential equations offer a rich and flexible modeling framework that has been applied to a large number of invasions. In this article, we are specifically interested in dating and localizing the introduction that led to an invasion using mathematical modeling, post-introduction data and an adequate statistical inference procedure. We adopt a mechanistic-statistical approach grounded on a coupled reaction-diffusion-absorption model representing the dynamics of an organism in an heterogeneous domain with respect to growth. Initial conditions (including the date and site of the introduction) and model parameters related to diffusion, reproduction and mortality are jointly estimated in the Bayesian framework by using an adaptive importance sampling algorithm. This framework is applied to the invasion of \textit{Xylella fastidiosa}, a phytopathogenic bacterium detected in South Corsica in 2015, France.

q-bio.PE↗