Pith. sign in

REVIEW 3 major objections 5 minor 68 references

Clustering of multivariate tail dependence using conditional methods

T0 review · 3 major / 5 minor · reviewed 2026-08-04 · deepseek-v4-flash

Pith's one-line read A closed-form divergence measure between conditional-extremes models makes clustering of multivariate tail dependence tractable in any dimension and beats bivariate-only methods.

desk verdict The method is a good idea, but the printed covariance formula makes the multivariate closed-form a scalar, so the central derivation needs fixing before the claims hold. read the letter →

arxiv 2510.20424 v2 pith:KOV2O4FB submitted 2025-10-23 stat.ME stat.AP

classification stat.MEstat.AP MSC 62G3262H30
keywords conditionalextremesmultivariatetaildependenceclusteringskew-geometricJensen-Shannondivergencek-medoidsextremevalueanalysiscompoundweatherevents
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper is trying to establish that the tail-dependence structure of a set of random vectors — for instance, precipitation and wind speed at many weather stations — can be grouped automatically by comparing their fitted conditional-extremes models, rather than by pairwise correlation coefficients alone. The proposed dissimilarity is the expected skew-geometric Jensen-Shannon divergence between the Gaussian conditional distributions implied by the conditional-extremes model, which has a closed form and is cheap to compute. Feeding these pair-wise divergences into a standard k-medoids algorithm yields clusters whose members share homogeneous joint tail behaviour, in arbitrary dimension. The paper argues this approach outperforms the existing nonparametric competitor in bivariate simulations and is the first to extend such clustering to more than two variables. If correct, it gives environmental and financial analysts a fast, model-based way to pool information across sites and to interpret spatially coherent extreme-event regimes.

What carries the argument

The central object is the expected conditional skew-geometric Jensen-Shannon divergence (eJSGλ). For a chosen conditioning variable i and two sites s and s*, it averages, over extreme values y of the conditioning variable, the skew-geometric Jensen-Shannon divergence between the two fitted Gaussian conditional distributions of the remaining d−1 variables given Yi = y. The skew-geometric Jensen-Shannon divergence is a symmetric, bounded, non-negative divergence built from the weighted geometric mean of two distributions; for Gaussians it reduces to a closed form in the precision-weighted mean and covariance. The expectation exploits the standard Laplace margins, whose exceedance density is id

What would settle it

Simulate data from conditional-extremes models whose residuals are deliberately non-Gaussian (e.g., skew-t or chi-square distributions) with known cluster labels, compute the Gaussian-assumption eJSGλ dissimilarities, and check whether clustering recovers the labels; if adjusted Rand index collapses while labels are still recoverable from the true conditional distributions, the divergence is tracking model misfit.

Watch

Extended reading notes

Core claim

On its own terms, the paper claims that the expected conditional skew-geometric Jensen-Shannon divergence between fitted conditional-extremes models is a one-number summary of multivariate tail dependence that preserves enough information for meaningful clustering. Under the standard working assumption that the residual vector in the conditional-extremes model is multivariate Gaussian, the divergence between two fitted conditional distributions has a closed-form expression, and the expectation over the conditioning variable above the threshold is computed by a cheap Monte Carlo integral. Clustering the resulting dissimilarity matrix with the Partitioning Around Medoids algorithm produces spa

Load-bearing premise

The closed-form divergence assumes the conditional-extremes model residuals are multivariate Gaussian; if the true conditional tails are non-Gaussian, the pairwise distances measure mismatch between fitted Gaussian approximations rather than difference between the actual tail distributions.

Editorial extensions

If this is right

  • Clustering first, pooling second: data from sites assigned to the same cluster can be pooled and the conditional-extremes model refitted, reducing parameter variance and bias, as the Gaussian-copula simulation demonstrates.
  • The method extends tail-dependence clustering to any dimension d, where the only existing benchmark is restricted to bivariate data; simulation shows accuracy rising with dimension up to d = 5.
  • The closed form keeps the method fast enough for realistic applications: 60 sites in simulation, 59 stations in the Irish application, with cluster assignments largely stable as the fitting threshold varies between the 0.85 and 0.90 quantiles.
  • Applied to weekly Irish precipitation and wind-speed data, the clusters define three spatially coherent regions — east, central, west — without any spatial information being used, and single out known outlier sites such as Malahide Castle.
  • The method handles both asymptotic dependence and asymptotic independence, unlike several existing extremal-clustering approaches that assume one regime only.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The load-bearing Gaussianity of the residuals is a modelling convenience, not a law: with misspecified residuals the closed form measures the divergence between fitted Gaussian approximations, and a numerical divergence computed from non-Gaussian fitted residuals would be the safer alternative in data with clearly skewed or heavy-tailed conditional tails.
  • The same exponential-family closure that makes the skew-geometric Jensen-Shannon divergence closed-form for Gaussians should extend to other exponential residual families, so the recipe generalizes beyond the particular working assumption used here.
  • Because the paper does not propagate parameter uncertainty into the dissimilarity matrix, cluster labels are point estimates; a bootstrap over fitted parameters would yield a distribution over clusterings, at a computational cost that the closed form was designed to avoid.
  • The claim of arbitrary dimension is established for d up to 5 with exchangeable pairwise dependence; the realistic high-dimensional case with heterogeneous, possibly sparse dependence across variable pairs is the natural next stress test.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The paper proposes a clustering method for multivariate extremes based on the conditional extremes (CE) framework of Heffernan and Tawn (2004). For each location/vector, a CE model is fitted for each conditioning variable under the working assumption of Gaussian residuals. The paper then defines a skew-geometric Jensen-Shannon (JSG) divergence between the fitted conditional Gaussian distributions, computes its expectation conditional on threshold exceedance, builds a pairwise dissimilarity matrix, and applies PAM clustering. Claims include a closed-form divergence, applicability in arbitrary dimensions, superior performance over Vignotto et al. (2021) in bivariate simulations, and a meaningful application to Irish precipitation/wind-speed data. The paper also reports extensions to d=3 and d=5 simulations and discusses limitations around the Gaussian residual assumption and uncertainty propagation.

Significance. If the multivariate derivation is correct, the method is a useful and computationally efficient tool for clustering tail-dependence structures, particularly because it handles both asymptotic dependence and asymptotic independence and avoids the bivariate restriction of existing nonparametric approaches. The paper is honest about the Gaussian working assumption and about the lack of uncertainty propagation, and the simulation studies are reasonably extensive. The main methodological novelty is the use of a closed-form divergence between fitted CE models to obtain a dissimilarity matrix. However, the central multivariate derivation as printed contains a dimensional error in the conditional covariance expression, which affects the claimed general-d applicability. The paper does not provide code or data, and the simulation evidence for d>2 cannot be checked against the printed equations.

major comments (3)
  1. [Section 2.3, Eq. (6) and Eq. (10)] Equation (6) states that the conditional distribution of Y_{-i,s} | Y_{i,s}=y is MVN with covariance matrix written as (y^{β_{|i,s}})^T Σ_{|i,s} y^{β_{|i,s}}. For d>2, β_{|i,s} is a (d-1)-vector and y^{β_{|i,s}} is also a (d-1)-vector, so this expression is a scalar, not a (d-1)×(d-1) matrix. Equation (10) then substitutes this scalar into the matrix formula (9), making the multivariate closed-form derivation invalid as printed. The bivariate case hides the problem because y^{β} is a scalar when d=2. The manuscript claims applicability in arbitrary dimensions, and the d=3 and d=5 simulations cannot be reproduced from Eq. (6). Please correct Eq. (6) to use diag(y^{β_{|i,s}}) Σ_{|i,s} diag(y^{β_{|i,s}}), and update Eq. (10) and the surrounding text accordingly. This is load-bearing for the paper's central claim.
  2. [Sections 2.3 and 3.2.2] The closed-form divergence is between fitted Gaussian conditional models, not directly between the true tail dependence structures. The paper acknowledges this, and the simulations show that clustering based on the fitted Gaussian models works for Gaussian and t-copula mixtures. However, the t-copula residuals are not Gaussian, so the simulation results indicate robustness rather than exactness. This is an acceptable limitation if clearly stated, but the phrase 'dissimilarity measure for multivariate tails' in the abstract should be tightened to 'dissimilarity between fitted CE models under the Gaussian working assumption', especially since the application and simulations all rely on that assumption.
  3. [Section 3.2.2 and Eq. (6)] The d=3 and d=5 simulation results are presented as evidence for multivariate applicability, but the printed equations do not specify how the covariance matrix is computed in those settings. The paper says the same ρ_t is used for all variable pairs, which makes the simulation favourable to aggregation across conditioning variables, but this does not resolve the dimensional inconsistency. Please either provide the corrected multivariate formula and a derivation, or include code so that the implementation can be verified.
minor comments (5)
  1. [Section 2.3, Eq. (7) vs Eq. (9)] The definition of JSGλ in Eq. (7) has no factor 1/2, while the closed-form formula in Eq. (9) has a leading 1/2. This is likely a convention difference, but it should be reconciled so the equations are consistent. Since a constant multiple does not affect clustering, this is not a substantive issue.
  2. [Section 3.2 and Figure caption 2] The text says 500 replicated experiments are performed, but Figure 2's caption refers to 'bootstrap samples'. Please make the terminology consistent; these are simulation replications, not bootstrap resamples.
  3. [Section 4.1 and Figure 6] The empirical χ(0.95) values are useful, but it would be clearer to report their uncertainty, even briefly. Many sites have very low χ estimates, and it is unclear whether those are distinguishable from zero.
  4. [Supplementary material] The supplement says 'Data and R code: Available upon request from the corresponding author.' For reproducibility, please deposit code and processed data in a public repository or add a detailed pseudo-code appendix, particularly because the multivariate divergence formula needs to be verified.
  5. [Section 2.4] The TWGSS elbow criterion is standard, but the paper says 'Unreported experiments showed that the TWGSS method correctly identified k' in Section 3. This is an unsupported claim; either report those experiments or soften the statement.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the divergence matrix is computed from fitted CE models and cluster accuracy is validated against external simulation truths.

full rationale

The paper's derivation chain is not circular. The dissimilarity in Eq. (10)-(11) is a function only of the fitted conditional Gaussian CE distributions in Eq. (6); no cluster assignment or validation label enters the construction of the eJSG divergence or of the dissimilarity matrix M in Eq. (12)-(13). Cluster labels are produced by PAM from M and are then compared, in Section 3, to the true generative clusters (Gaussian-copula correlations or t-copula correlation groups) via the ARI, i.e. an external, data-generating benchmark; the comparison to Vignotto et al. (2021) is likewise an external competitor. The Gaussian working assumption is disclosed in Section 2.2 ('we follow Heffernan and Tawn (2004) and assume that the residual vector ... is multivariate Gaussian') and its limitations are explicitly acknowledged in Section 5 ('While the residual Gaussian assumption on which our method relies may not hold for all datasets' and 'We do not propagate uncertainty in the CE model estimates through to the clustering'); assumption misspecification and unpropagated uncertainty are robustness limitations, not circular reductions. Self-citations to Rohrbeck and Tawn (2021), Richards et al. (2022, 2023) and Talento et al. (2025) are contextual/related-work citations and do not carry the argument. The closed-form JSG formula is attributed to Nielsen (2019), an external mathematical result, and no uniqueness claim is imported from the authors' own prior work. The only caveat close to circularity is that the Gaussian-copula simulation is partly in-sample for the Gaussian working assumption, but because the method is also tested on t-copula mixtures and compared with an external method, this does not make the central claim equivalent to its inputs. The statement 'Unreported experiments showed that the TWGSS method correctly identified k' is an omitted-support concern, not circularity. A dimensional inconsistency in Eq. (6) for d>2 would be a correctness issue, not circularity.

Assumptions & free parameters 4 free parameters · 3 assumptions · 0 invented entities

The central method rests on the CE asymptotic approximation and the Gaussian residual working assumption, plus several tuning choices (threshold, lambda, k, truncation). No new physical or probabilistic entity is postulated.

free parameters (4)
  • Conditioning threshold quantile q = 0.85 in application; 0.9/0.99 in simulations
    Selected via threshold stability plots; affects which exceedances define the tail and hence the fitted CE parameters.
  • Skew parameter lambda for JSG = 0.5
    Fixed to make the divergence symmetric; a user choice that could weight sites differently.
  • Number of clusters k = 3 in application; true k in simulations
    Chosen by TWGSS elbow; central to the clustering output.
  • Upper truncation quantile for MC integral = 0.99 of pooled empirical distribution
    Truncates the integral in Eq (11) to avoid extrapolation; an ad hoc choice affecting the expected divergence.
assumptions (3)
  • domain assumption CE asymptotic limit in Eq (3) holds exactly above a finite threshold u
    Standard EVT approximation used to justify the regression form in Eq (4); its accuracy depends on threshold choice.
  • domain assumption Residual vector Z|i,s is multivariate Gaussian
    Working assumption introduced in Section 2.2 to enable a closed-form divergence; acknowledged as a limitation in Section 5.
  • domain assumption All sites share standard Laplace marginal distributions and the same threshold u_i, giving identical conditional densities h(y | Y_i,s > u_i)
    Required for symmetry of eJSG in Eq (11); relies on the probability integral transform to Laplace margins.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Clustering of multivariate tail dependence using conditional methods." pith.science (2026). https://pith.science/paper/KOV2O4FB

@misc{pith2026251020424,
  author       = {Pith},
  title        = {Pith review of: Clustering of multivariate tail dependence using conditional methods},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KOV2O4FB}},
  note         = {Machine review of arXiv:2510.20424}
}
read the original abstract

The conditional extremes (CE) framework has proven useful for analysing the joint tail behaviour of random vectors. However, when applied across many locations or variables, it can be difficult to interpret or compare the resulting extremal dependence structures, particularly for high dimensional vectors. To address this, we propose a novel clustering method for multivariate extremes using the CE framework. Our approach introduces a closed-form, computationally efficient dissimilarity measure for multivariate tails, based on the skew-geometric Jensen-Shannon divergence, and is applicable in arbitrary dimensions. Applying standard clustering algorithms to a matrix of pairwise distances, we obtain interpretable groups of random vectors with homogeneous tail dependence. Simulation studies demonstrate that our method outperforms existing approaches for clustering bivariate extremes, and uniquely extends to the multivariate setting. In our application to Irish meteorological data, our clustering identifies spatially coherent regions with similar extremal dependence between precipitation and wind speeds.

Figures

Figures reproduced from arXiv: 2510.20424 by the authors.

Figure 1
Figure 1. shows boxplots of the estimated model parameters stratified by the true values of ρ s Gauss. The results are shown pre- and post-clustering, with the latter derived by pooling data across all sites according to their estimated clustering (which was always correct in this setting). By pooling data across sites, we reduce estimation uncertainty for both α|i and β|i . Bias, as defined as the difference between the fini… view at source ↗
Figure 2
Figure 2. Comparison of our conditional extremes-based approach and the method of Vignotto et al. (2021). The x-axis represents the Gaussian correlation parameter ρGauss, with facet labels indicating the t-copula correlation parameters, ρt1 and ρt2 , for the two true clusters. The smoothed lines show the average ARI across bootstrap samples, for both methods, with smoothing performed using a generalised additive model. 3.2.2 … view at source ↗
Figure 3
Figure 3. Comparison of clustering performance for d = 2 and d = 3 variables and two equally-sized clusters. The x-axis represents the Gaussian correlation parameter ρGauss for both clusters, and the facet labels show the t-copula correlation parameters, ρt1 and ρt2 , for the two true clusters. The smoothed lines show the average ARI across bootstrap samples, with smoothing performed using a generalised additive model. To fur… view at source ↗
Figures from the paper (7 more)
Figure 4
Figure 4. Figure 4: Evaluation of clustering for d = 2 to d = 5 variables and two equally-sized clusters for simulations at 12 sites from a mixture of Gaussian and t-copulas. The t-copula correlation parameters for each true cluster were set to ρt1 = 0.4 and ρt2 = 0.5. A sequence of value…
Figure 5
Figure 5. Figure 5: Evaluation of clustering performance for d = 2 variables and three clusters of 60 locations for simulations from a mixture of Gaussian and t-copulas. The true x-axis and facet labels show the t-copula correlation parameters, ρt1 , ρt2 , and ρt3 , for each of the three …
Figure 6
Figure 6. Figure 6: Left: elevation (m) profile of Ireland. Right: Estimated χ(0.95) between precipi￾tation and wind speed at all 59 locations. Darker colours and larger points indicate stronger asymptotic dependence. Locations in red are (1) Malahide Castle, Dublin, (2) Ringsend, Dublin,…
Figure 7
Figure 7. Figure 7: Maps of site-wise estimated model parameters α|i,s (top) and β|i,s (bottom) for precipitation conditional on wind speed (left) and wind speed conditional on precipitation (right). The colour scale is the same for both variables, with darker red and larger points indica…
Figure 8
Figure 8. Figure 8: Extremal cluster estimates using the dissimilarity matrix for wind speed conditional on precipitation (centre), precipitation conditional on wind speed (right), and the combined dissimilarity matrix (left). Sites are coloured by cluster label. the difference between th…
Figure 9
Figure 9. Figure 9: Heatmap of the estimated aggregated dissimilarity matrix M. The names of each site are coloured and ordered according to their estimated cluster labels, matching that of [PITH_FULL_IMAGE:figures/full_fig_p026_9.png]
Figure 10
Figure 10. Figure 10: Estimated extremal clusters with the conditional extremes model parameters estimated at q = 0.85, q = 0.88 and q = 0.9 quantiles, using the aggregated dissimilarity matrix. Sites are coloured by cluster label. q = 0.90. Apart from these observations, the estimated clu…

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

68 extracted references · 3 linked inside Pith

  1. [1]

    M., de Zea Bermudez, P., and Scotto, M

    Alonso, A. M., de Zea Bermudez, P., and Scotto, M. G. (2014). Comparing generalized Pareto models fitted to extreme observations: an application to the largest temperatures in Spain . Stochastic Environmental Research and Risk Assessment , 28(5):1221--1233

  2. [2]

    I., and Keenlyside, N

    Bengtsson, L., Hodges, K. I., and Keenlyside, N. (2009). Will extratropical storms intensify in a warmer climate? Journal of Climate , 22(9):2276--2301

  3. [3]

    Bernard, E., Naveau, P., Vrac, M., and Mestre, O. (2013). Clustering of maxima: Spatial dependencies among heavy rainfall in France . Journal of Climate , 26(20):7929--7937

  4. [4]

    F., Ramos, A

    Bevacqua, E., De Michele, C., Manning, C., Couasnon, A., Ribeiro, A. F., Ramos, A. M., Vignotto, E., Bastos, A., Blesi \'c , S., Durante, F., et al. (2021). Guidelines for studying diverse types of compound weather and climate events. Earth's Future , 9(11):e2021EF002340

  5. [5]

    Boulin, A., Di Bernardino, E., Lalo \"e , T., and Toulemonde, G. (2025). Identifying regions of concomitant compound precipitation and wind speed extremes over Europe . Journal of the Royal Statistical Society Series C: Applied Statistics . qlaf014

  6. [6]

    Carreau, J., Naveau, P., and Neppel, L. (2017). Partitioning into hazard subregions for regional peaks-over-threshold modeling of heavy precipitation. Water Resources Research , 53(5):4407--4426

  7. [7]

    Coles, S. (2001). An Introduction to Statistical Modeling of Extreme Values . Springer Series in Statistics. Springer

  8. [8]

    Coles, S., Heffernan, J., and Tawn, J. (1999). Dependence measures for extreme value analyses. Extremes , 2(4):339--365

Show all 68 references
  1. [9]

    Cooley, D., Naveau, P., and Poncet, P. (2006). Variograms for Spatial Max--Stable Random Fields . In Dependence in Probability and Statistics , pages 373--390. Springer, New York

  2. [10]

    ERA5 hourly data on single levels from 1940 to present

    Copernicus Climate Change Service (C3S) (2024). ERA5 hourly data on single levels from 1940 to present . Copernicus Climate Change Service (C3S) Climate Data Store (CDS). Available at https://cds.climate.copernicus.eu/, last access: 31 January 2025

  3. [11]

    Davison, A. C. and Huser, R. (2015). Statistics of extremes. Annual Review of Statistics and Its Application , 2(1):203--235

  4. [12]

    Dawkins, L. C. and Stephenson, D. B. (2018). Quantification of extremal dependence in spatial natural hazard footprints: independence of windstorm gust speeds and its impact on aggregate losses. Natural Hazards and Earth System Sciences , 18(11):2933--2949

  5. [13]

    de Carvalho, M., Huser, R., and Rubio, R. (2023). Similarity-based clustering for patterns of extreme values. Stat , 12(1):e560

  6. [14]

    Deasy, J., Simidjievski, N., and Li \`o , P. (2020). Constraining Variational Inference with Geometric Jensen–Shannon Divergence . Advances in Neural Information Processing Systems , 33:10647--10658

  7. [15]

    and McNeil, A

    Demarta, S. and McNeil, A. J. (2007). The t Copula and Related Copulas . International Statistical Review , 73(1):111--129

  8. [16]

    Duchi, J. (2007). Derivations for Linear Algebra and Optimization . http://ai.stanford.edu/ jduchi/projects/general_notes.pdf. Accessed: 2025-01-31

  9. [17]

    J., Engelke, S., and Trapin, L

    Dupuis, D. J., Engelke, S., and Trapin, L. (2023). Modeling panels of extremes. The Annals of Applied Statistics , 17(1):498--517

  10. [18]

    F., and Lee, C

    Farrell, A., Eastoe, E. F., and Lee, C. (2025). Conditional Extremes with Graphical Models . arXiv preprint arXiv:2411.17013

  11. [19]

    French, J., Kokoszka, P., Stoev, S., and Hall, L. (2019). Quantifying the risk of heat waves using extreme value theory and spatio-temporal functional data. Computational Statistics & Data Analysis , 131:176--193

  12. [20]

    G., and Ammann, C

    Gilleland, E., Brown, B. G., and Ammann, C. M. (2013). Spatial extreme value analysis to project extremes of large-scale indicators for severe weather. Environmetrics , 24(6):418--432

  13. [21]

    B., Huser, R., and Ombao, H

    Guerrero, M. B., Huser, R., and Ombao, H. (2023). Conex--connect: Learning patterns in extremal brain connectivity from multichannel eeg data. The Annals of Applied Statistics , 17(1):178--198

  14. [22]

    Hastie, T., Tibshirani, R., and Friedman, J. (2009). The Elements of Statistical Learning . Springer Series in Statistics. Springer, New York

  15. [23]

    Heffernan, J. E. and Tawn, J. A. (2004). A conditional approach for multivariate extreme values (with discussion). Journal of the Royal Statistical Society Series B: Statistical Methodology , 66(3):497--546

  16. [24]

    Hilal, S., Poon, S.-H., and Tawn, J. (2014). Portfolio risk assessment using multivariate extreme value methods. Extremes , 17(4):531--556

  17. [25]

    and Arabie, P

    Hubert, L. and Arabie, P. (1985). Comparing partitions. Journal of Classification , 2(1):193--218

  18. [26]

    Huser, R., Opitz, T., and Wadsworth, J. (2024). Modeling of spatial extremes in environmental data science: Time to move away from max-stable processes

  19. [27]

    Jane, R., Cadavid, L., Obeysekera, J., and Wahl, T. (2020). Multivariate statistical modelling of the drivers of compound flood events in south Florida . Natural Hazards and Earth System Sciences , 20(10):2681--2699

  20. [28]

    Kakampakou, L., Simpson, E., and Wadsworth, J. (2024). Spatial Extremal Modelling : A Case Study on the Interplay Between Margins and Dependence . Stat , 13(4)

  21. [29]

    Keef, C., Papastathopoulos, I., and Tawn, J. A. (2013). Estimation of the conditional distribution of a multivariate variable given that one of its components is large: Additional constraints for the Heffernan and Tawn model . Journal of Multivariate Analysis , 115:396–404

  22. [30]

    and Opitz, T

    Koh, J. and Opitz, T. (2025). Extreme-value modelling of migratory bird arrival dates: Insights from citizen science data (with discussion). Journal of the Royal Statistical Society Series A: Statistics in Society , 188(3):674--699

  23. [31]

    Koh, J., Pimont, F., Dupuy, J.-L., and Opitz, T. (2023). Spatiotemporal wildfire modeling through point processes with moderate and extreme marks. The Annals of Applied Statistics , 17(1):560--582

  24. [32]

    and Leibler, R

    Kullback, S. and Leibler, R. A. (1951). On information and sufficiency. The Annals of Mathematical Statistics , 22(1):79--86

  25. [33]

    Lin, J. (1991). Divergence Measures Based on the Shannon Entropy . IEEE Transactions on Information Theory , 37(1):145--151

  26. [34]

    Historical Data - Met Éireann - The Irish Meteorological Service --- met.ie

    Met Éireann (2025). Historical Data - Met Éireann - The Irish Meteorological Service --- met.ie . https://www.met.ie/climate/available-data/historical-data. [Accessed 2025-08-01]

  27. [35]

    and Davison, A

    Miralles, O. and Davison, A. C. (2024). Bayesian modeling of insurance claims for hail damage. The Annals of Applied Statistics , 18(4):3091--3108

  28. [36]

    Mornet, A., Opitz, T., Luzi, M., Loisel, S., and Bailleul, B. (2017). Wind storm risk management: sensitivity of return period calculations and spread on the territory. Stochastic Environmental Research and Risk Assessment , 31(8):1977--1995

  29. [37]

    Ni, S., Lin, C., Wang, H., Li, Y., Liao, Y., and Li, N. (2023). Learning geometric Jensen-Shannon divergence for tiny object detection in remote sensing images. Frontiers in Neurorobotics , 17:1273251

  30. [38]

    Nielsen, F. (2019). On the Jensen-Shannon Symmetrization of Distances Relying on Abstract Means . Entropy , 21(5):485

  31. [39]

    Nielsen, F. (2025). Two tales for a geometric Jensen-Shannon divergence . arXiv preprint arXiv:2508.05066

  32. [40]

    Opitz, T. (2016). Modeling asymptotically independent spatial extremes based on Laplace random fields. Spatial Statistics , 16:1--18

  33. [41]

    A., Amekudzi, L

    Osei, M. A., Amekudzi, L. K., Omari-Sasu, A. Y., Yamba, E. I., Quansah, E., Aryee, J. N., and Preko, K. (2021). Estimation of the return periods of maxima rainfall and floods at the Pra River Catchment, Ghana, West Africa using the Gumbel extreme value theory. Heliyon , 7(5)

  34. [42]

    Poon, S.-H., Rockinger, M., and Tawn, J. (2004). Extreme value dependence in financial markets: Diagnostics , models, and financial implications. The Review of Financial Studies , 17(2):581--610

  35. [43]

    B., Redondo, P., and Shao, X

    Richards, J., Alotaibi, N., Cisneros, D., Gong, Y., Guerrero, M. B., Redondo, P., and Shao, X. (2023). Modern extreme value statistics for Utopian extremes. arXiv preprint arXiv:2311.11054

  36. [44]

    A., and Brown, S

    Richards, J., Tawn, J. A., and Brown, S. (2022). Modelling extremes of spatial aggregates of precipitation using conditional methods. The Annals of Applied Statistics , 16(4):2693--2713

  37. [45]

    F., Frigessi, A., and Tawn, J

    Rohrbeck, C., Eastoe, E. F., Frigessi, A., and Tawn, J. A. (2018). Extreme value modelling of water-related insurance claims. The Annals of Applied Statistics , 12(1):246--282

  38. [46]

    and Tawn, J

    Rohrbeck, C. and Tawn, J. A. (2021). Bayesian spatial clustering of extremal behavior for hydrological variables. Journal of Computational and Graphical Statistics , 30(1):91--105

  39. [47]

    and Kaufman, P

    Rousseeuw, L. and Kaufman, P. (1987). Clustering by means of medoids. In Statistical Data Analysis Based on the L1 Norm and Related Methods , pages 405--416. Neuchâtel, Switzerland

  40. [48]

    Sando, K., Wada, R., Rohmer, J., Lecacheux, S., and Jonathan, P. (2022). Estimating Joint Extremes of Significant Wave Height and Wind Speed for Tropical Cyclones . In Volume 5B: Ocean Engineering; Honoring Symposium for Professor G\" u nther F. Clauss on Hydrodynamics and Oce...

  41. [49]

    Shao, X., Hazra, A., Richards, J., and Huser, R. (2025). Flexible modeling of nonstationary extremal dependence using spatially fused LASSO and ridge penalties. Technometrics , 67(1):97--111

  42. [50]

    S., Opitz, T., and Wadsworth, J

    Simpson, E. S., Opitz, T., and Wadsworth, J. L. (2023). High-dimensional modeling of spatial and spatio-temporal conditional extremes using INLA and Gaussian Markov random fields . Extremes , 26(4):669–713

  43. [51]

    Simpson, E. S. and Wadsworth, J. L. (2021). Conditional modelling of spatio-temporal extremes for Red Sea surface temperatures. Spatial Statistics , 41:100482

  44. [52]

    Soukissian, T. H. and Tsalis, C. (2015). The effect of the generalized extreme value distribution parameter estimation methods in extreme wind speed prediction. Natural Hazards , 78:1777--1809

  45. [53]

    and Heffernan, J

    Southworth, H. and Heffernan, J. E. (2012). Multivariate extreme value modelling of laboratory safety data from clinical studies. Pharmaceutical Statistics , 11(5):367--372

  46. [54]

    A., and Kl \"u ppelberg, C

    Steinkohl, C., Davis, R. A., and Kl \"u ppelberg, C. (2013). Extreme value analysis of multivariate high-frequency wind speed data. Journal of Statistical Theory and Practice , 7:73--94

  47. [55]

    Talento, M. S. D., Richards, J., Pinto-Orellana, M., Huser, R., and Ombao, H. C. (2025). Spectral Extremal Connectivity of Two-State Seizure Brain Waves . arXiv preprint arXiv:2503.04169

  48. [56]

    Tanarhte, M., Hadjinicolaou, P., and Lelieveld, J. (2015). Heat wave characteristics in the eastern Mediterranean and Middle East using extreme value theory. Climate Research , 63(2):99--113

  49. [57]

    Tendijck, S., Eastoe, E., Tawn, J., Randell, D., and Jonathan, P. (2023). Modeling the Extremes of Bivariate Mixture Distributions With Application to Oceanographic Data . Journal of the American Statistical Association , 118(542):1373--1384

  50. [58]

    and Ghosh, S

    Thiagarajan, P. and Ghosh, S. (2025). Jensen-Shannon divergence based novel loss functions for Bayesian neural networks. Neurocomputing , 618(129115):129115

  51. [59]

    Thorndike, R. L. (1953). Who belongs in the family? Psychometrika , 18(4):267--276

  52. [60]

    Vettori, S., Huser, R., and Genton, M. G. (2019). Bayesian modeling of air pollution extremes using nested multivariate max-stable processes. Biometrics , 75(3):831--841

  53. [61]

    Vignotto, E., Engelke, S., and Zscheischler, J. (2021). Clustering bivariate dependencies of compound precipitation and wind extremes over Great Britain and Ireland . Weather and Climate Extremes , 32:100318

  54. [62]

    and Laxman, S

    Vijendra, S. and Laxman, S. (2015). Symmetry based automatic evolution of clusters: A new approach to data clustering. Computational Intelligence and Neuroscience , 2015:796276

  55. [63]

    and Tawn, J

    Wadsworth, J. and Tawn, J. (2022). Higher-dimensional spatial extremes via single-site conditioning. Spatial Statistics , 51:100677

  56. [64]

    C., Tawn, J

    Winter, H. C., Tawn, J. A., and Brown, S. J. (2016). Modelling the effect of the El Niño-Southern Oscillation on extreme spatial temperature events over Australia . The Annals of Applied Statistics , 10(4):2075 -- 2101

  57. [65]

    D., Wehner, M

    Zhang, L., Risser, M. D., Wehner, M. F., and O'Brien, T. A. (2024). Leveraging Extremal Dependence to Better Characterize the 2021 Pacific Northwest Heatwave . Journal of Agricultural, Biological and Environmental Statistics

  58. [66]

    Zheng, S., Fan, K., Hou, Y., Feng, J., and Fu, Y. (2023). Clustering by the probability distributions from extreme value theory. IEEE Transactions on Artificial Intelligence , 4(2):292--303

  59. [67]

    Zscheischler, J., Naveau, P., Martius, O., Engelke, S., and Raible, C. C. (2021). Evaluating the dependence structure of compound precipitation and wind speed extremes. Earth System Dynamics , 12(1):1--16

  60. [68]

    J., Seneviratne, S

    Zscheischler, J., Westra, S., Van Den Hurk, B. J., Seneviratne, S. I., Ward, P. J., Pitman, A., AghaKouchak, A., Bresch, D. N., Leonard, M., Wahl, T., et al. (2018). Future climate risk from compound events. Nature Climate Change , 8(6):469--477

Pith tools

Reviewed August 4, 2026 · model on record in the stance chip above.