arXiv subjects

Chenglei Hu

Publications and source records attributed to Chenglei Hu.

3 recordsLinked to original sources

arXiv2025

XGBoost meets INLA: a two-stage spatio-temporal forecasting of wildfires in Portugal

Wildfires pose a major threat to Portugal, with over 115,000 hectares burned annually on average during 1980-2024, and the country has faced devastating mega-fires such as those in 2017. Accurate forecasts of wildfire occurrence and burned area are therefore essential for firefighting resource allocation and emergency preparedness. In this study, we propose a novel two-stage ensemble that extends the widely used latent Gaussian modelling framework with integrated nested Laplace approximation (INLA) for spatio-temporal wildfire forecasting. Stage 1 applies a gradient boosting model (XGBoost) to environmental covariates and historical fire records to produce one-month-ahead point forecasts of fire counts and burned area. Stage 2 uses these predictions as external covariates in a latent Gaussian model with additional spatiotemporal random effects to generate probabilistic forecasts of monthly total fire counts and burned area at the council level. To capture both moderate and extreme events, we implement the extended generalised Pareto (eGP) likelihood (a sub-asymptotic distribution) within INLA, develop Penalised Complexity (PC) priors for its parameters, and compare the eGP likelihood with common alternatives (e.g., Gamma and Weibull). Our framework tackles the unavailability of future environmental covariates at prediction time and performs strongly for one-month-ahead forecasts.

stat.AP↗

arXiv2025

GPDFlow: Generative Multivariate Threshold Exceedance Modeling via Normalizing Flows

The multivariate generalized Pareto distribution (mGPD) is a common method for modeling extreme threshold exceedance probabilities in environmental and financial risk management. Despite its broad applicability, mGPD faces challenges due to the infinite possible parametrizations of its dependence function, with only a few parametric models available in practice. To address this limitation, we introduce GPDFlow, an innovative mGPD model that leverages normalizing flows to flexibly represent the dependence structure. Unlike traditional parametric mGPD approaches, GPDFlow does not impose explicit parametric assumptions on dependence, resulting in greater flexibility and enhanced performance. Additionally, GPDFlow allows direct inference of marginal parameters, providing insights into marginal tail behavior. We derive tail dependence coefficients for GPDFlow, including a bivariate formulation, a $d$-dimensional extension, and an alternative measure for partial exceedance dependence. A general relationship between the bivariate tail dependence coefficient and the generative samples from normalizing flows is discussed. Through simulations and a practical application analyzing the risk among five major US banks, we demonstrate that GPDFlow significantly improves modeling accuracy and flexibility compared to traditional parametric methods.

stat.ME↗

arXiv2024

A Bayesian multivariate extreme value mixture model

Impact assessment of natural hazards requires the consideration of both extreme and non-extreme events. Extensive research has been conducted on the joint modeling of bulk and tail in univariate settings; however, the corresponding body of research in the context of multivariate analysis is comparatively scant. This study extends the univariate joint modeling of bulk and tail to the multivariate framework. Specifically, it pertains to cases where multivariate observations exceed a high threshold in at least one component. We propose a multivariate mixture model that assumes a parametric model to capture the bulk of the distribution, which is in the max-domain of attraction (MDA) of a multivariate extreme value distribution (mGEVD). The tail is described by the multivariate generalized Pareto distribution, which is asymptotically justified to model multivariate threshold exceedances. We show that if all components exceed the threshold, our mixture model is in the MDA of an mGEVD. Bayesian inference based on multivariate random-walk Metropolis-Hastings and the automated factor slice sampler allows us to incorporate uncertainty from the threshold selection easily. Due to computational limitations, simulations and data applications are provided for dimension $d=2$, but a discussion is provided with views toward scalability based on pairwise likelihood.

stat.ME↗