New frontiers in Bayesian modeling using the INLA package in R

Niekerk, Janet van; Bakka, Haakon; Rue, Haåvard; Schenk, Olaf

doi:10.48550/arxiv.1907.10426

Cited by 10 publications

(10 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the study, we considered the heterogeneity within each train in both two statistical models, however, the heterogeneity among trains are not touched yet and could be considered in the further investigation, for example frailty Cox model and/or fitting the two models from Bayesian perspective with random effects among trains (Niekerk et al, 2019). In addition, how to choose the changing point and how many changing points in a heterogeneous Markov chain process become critical problems, since the estimated transition intensity matrix may be sensitive to the choices which are very subjective.…”

Section: Conclusion and Discussionmentioning

confidence: 99%

Train performance analysis using heterogeneous statistical models

Wang¹,

Yu²

2021

Preprint

View full text Add to dashboard Cite

This study investigated the effect of harsh winter climate on the performance of high speed passenger trains in northern Sweden. Novel approaches based on heterogeneous statistical models were introduced to analyse the train performance in order to take the time-varying risks of train delays into consideration. Specifically, stratified Cox model and heterogeneous Markov chain model were used for modelling primary delays and arrival delays, respectively. Our results showed that the weather variables including temperature, humidity, snow depth, and ice/snow precipitation have significant impact on the train performance.

show abstract

Section: Conclusion and Discussionmentioning

confidence: 99%

Train performance analysis using heterogeneous statistical models

Wang¹,

Yu²

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Stable inferences may require compromises with respect to the complexity of the latent model and the number of observations, which jointly determine the size and sparsity of the Gaussian precision matrices, which in turn influence computation times, memory requirements and well-conditioned numerical behavior. Even stronger restrictions arise with methods such as Markov Chain Monte Carlo (MCMC) to achieve approximation quality comparable to INLA (Taylor & Diggle 2014, van Niekerk et al 2019. Krainski et al (2018, §8.4) develop strategies for LGCPs by aggregating the events to larger mapping units and lowering spatial-temporal resolution of random effects to decrease computation times, which, however, would impede modeling structures arising at small spatiotemporal scales.…”

Section: Data Aggregation and Subsampling Schemesmentioning

confidence: 99%

Spatiotemporal wildfire modeling through point processes with moderate and extreme marks

Koh¹,

Pimont²,

Dupuy³

et al. 2021

Preprint

View full text Add to dashboard Cite

Accurate spatiotemporal modeling of conditions leading to moderate and large wildfires provides better understanding of mechanisms driving fire-prone ecosystems and improves risk management. We study daily summer wildfire data for the French Mediterranean basin during 1995-2018. We jointly model occurrence intensity and wildfire sizes by combining extreme-value theory and point processes within a novel Bayesian hierarchical model. The occurrence component models wildfire ignitions as a spatiotemporal log-Gaussian Cox process. Burnt areas are numerical marks attached to points, and we consider fires with marks exceeding a high threshold as extreme. The size component is a two-component mixture varying in space and time that jointly models moderate and extreme fires. We capture non-linear influence of covariates (Fire Weather Index, forest cover) through component-specific smooth functions, which may vary with season. To reveal common drivers of different aspects of wildfire activity, we share random effects between model components to improve interpretability and parsimony without compromising predictive skill. Stratified subsampling of zero counts is implemented to cope with large observation vectors. We compare and validate models through predictive scores and visual diagnostics. Our methodology provides a holistic approach to explaining and predicting the drivers of wildfire activity and their associated uncertainties.

show abstract

“…The integrated nested Laplace approximation (INLA Rue et al, 2009, 2017Opitz, 2017;van Niekerk et al, 2019) provides relatively fast and accurate analytical approximations for posterior inference in models with latent Gaussian processes. The distribution of the observed variables may be non-Gaussian conditional on the latent Gaussian process, although here the focus of our modeling approach for conditional extremes will be on Gaussian responses;…”

Section: Bayesian Inference With the Integrated Nested Laplace Approx...mentioning

confidence: 99%

High-dimensional modeling of spatial and spatio-temporal conditional extremes using INLA and Gaussian Markov random fields

Simpson¹,

Opitz²,

Wadsworth³

2020

Preprint

View full text Add to dashboard Cite

The conditional extremes framework allows for event-based stochastic modeling of dependent extremes, and has recently been extended to spatial and spatio-temporal settings. After standardizing the marginal distributions and applying an appropriate linear normalization, certain non-stationary Gaussian processes can be used as asymptotically-motivated models for the process conditioned on threshold exceedances at a fixed reference location and time. In this work, we adopt a Bayesian perspective by implementing estimation through the integrated nested Laplace approximation (INLA), allowing for novel and flexible semi-parametric specifications of the Gaussian mean function. By using Gauss-Markov approximations of the Matérn covariance function (known as the Stochastic Partial Differential Equation approach) at a latent stage of the model, likelihood-based inference becomes feasible even with several thousands of observed locations. We explain how constraints on the spatial and spatio-temporal Gaussian processes, arising from the conditioning mechanism, can be implemented through the latent variable approach without losing the computationally convenient Markov property. We discuss tools for the comparison of posterior models, and illustrate the flexibility of the approach with gridded Red Sea surface temperature data at over 6, 000 observed locations. Posterior sampling is exploited to study the probability distribution of cluster functionals of spatial and spatio-temporal extreme episodes.

show abstract

New frontiers in Bayesian modeling using the INLA package in R

Cited by 10 publications

References 37 publications

Train performance analysis using heterogeneous statistical models

Train performance analysis using heterogeneous statistical models

Spatiotemporal wildfire modeling through point processes with moderate and extreme marks

High-dimensional modeling of spatial and spatio-temporal conditional extremes using INLA and Gaussian Markov random fields

Contact Info

Product

Resources

About