Coverage of academic papers and research directions relevant to UK insurance pricing. What the literature says, what it omits, and whether the methods survive contact with messy real-world data. 6 articles.
A February 2026 paper provides the first statistically valid confidence intervals for global SHAP feature importance. We explain what changes for UK insurance pricing teams, whe...
Liu & Meng's PowerBurr adds a fourth parameter to Burr XII that decouples body shape from tail heaviness. We explain the coupling trap, what the fix actually does, and when UK p...
Boucher & Coulibaly (arXiv:2502.11788) prove that offset and ratio exposure handling are equivalent for Poisson frequency models — but diverge for Tweedie pure-premium GLMs, whe...
Negative log-likelihood is a proper scoring rule. So why does NLL training collapse a Mixture Density Network to a single component? The answer is in the loss surface geometry, ...
A new 173-page handbook from MARSAIL documents one Thai insurer's computer vision system for vehicle damage assessment. It contains one genuinely useful data point — 50% VIN acc...
IJCNN 2025 paper arXiv:2509.00846 introduces Causal SHAP: it uses causal discovery to estimate a DAG, then computes SHAP values that respect causal structure. The correlated-fea...
Nieto-Barajas (arXiv:2602.07228) proposes a Bayesian nonparametric mixture of Shifted Gamma-Gamma distributions that eliminates EVT threshold selection entirely and produces a p...
Orihara, Momozaki & Sugasawa (arXiv:2506.04868) produce a Bayesian posterior over the ATE by tilting the product of independent posteriors to satisfy the DR moment condition. We...
Wang & Yu (arXiv:2603.01232) characterise which risk measures are submodular — mathematically encoding that diversification benefits are real but bounded. ES is submodular; VaR ...
Das (arXiv:2603.24640) establishes stochastic ordering results for minimum and maximum claims across heterogeneous portfolios with random claim counts. Useful theoretical scaffo...
Boonen & Ghossoub (arXiv:2602.14223) prove that when a reinsurer sets terms first — Bowley/Stackelberg — the treaty is mathematically guaranteed to be worse for total welfare th...
A new paper models hazard functions as solutions to nonlinear ODEs, producing shapes no standard parametric family can match. The maths is genuinely interesting. The absence of ...
Yang et al. (arXiv:2603.27672) fix mode collapse in Mixture Density Networks by adding an analytic Energy Score term to the training objective. The contribution is real and spec...
Samuel Clark's MDMx (arXiv:2603.20518) applies Tucker tensor decomposition to HMD mortality data across 50 countries, producing coherent forecasts, disruption detection, and mod...
When you acquire a portfolio or enter a scheme, your pricing model was fitted on a different risk population. Weill and Wang (2026) give a kernel GLM framework for correcting th...
Shankar & Cohen automate GAM structure search using NSGA-II evolutionary algorithms. The idea is legitimate; the problem is that EBM already does this better for insurance prici...
Asadi & Li's Generative Adversarial Regression (arXiv:2603.08553) frames scenario generation as a minimax problem between a generator and an adversarial policy. For Solvency II ...
A new paper (arXiv:2504.09396) uses PPO reinforcement learning with a CVaR constraint to manage a reserve buffer. The framing is interesting — but this is not reserving in the a...
Nieto-Barajas (arXiv:2602.07228, 2026) proposes a Bayesian nonparametric mixture of Scaled Generalised Gaussian distributions that eliminates threshold selection entirely. The m...
Malandii & Uryasev's Biased Mean Quadrangle (arXiv:2603.26901) provides a linear-programming-based method for estimating E[Y]+x — a biased mean offset. For reserving actuaries, ...
Laub/Pho/Wong's Actuarial Neural Additive Model has a genuine architectural insight in PWLCalibration monotonicity. It also depends on an unmaintained TensorFlow library. EBM is...
An honest assessment of where tabular foundation models stand in March 2026 — what the benchmarks actually show, what's missing for insurance pricing, and which models are worth...
Deep learning survival models underperform Cox regression on tabular insurance data. Cure models are the real story post-GIPP. Here is what the research says and what UK pricing...
A practitioner-oriented deep dive on applying reinforcement learning and contextual bandits to PCW margin optimisation for UK personal lines. Two serious papers exist, six hard ...
Theory proves standard XL layers are optimal under VaR. Yet most pricing teams set retentions by trial and error. We review the academic results, the commercial tools, and the P...
Richman and Wüthrich's March 2026 paper shows linear regression with projection-to-ultimate factors closes 44% of the gap over chain ladder — and neural networks add nothing at ...
Miao & Pesenti (arXiv:2603.16720) derive discrimination-insensitive premiums by finding the probability measure nearest to the real-world measure in KL-divergence, subject to ze...
A stochastic SIR model calibrated to LockBit ransomware data shows why treating cyber losses as independent events badly underestimates portfolio-level risk.
Balzer and Benlahlou (arXiv:2603.14543) extend gradient boosting to spatial panel data. Here is what it does, how it compares to BYM2 and Blier-Wong, and when a UK pricing team ...
Samuel Clark's MDMx organises the Human Mortality Database as a four-way tensor and applies Tucker decomposition to produce structurally coherent mortality models across 50 coun...
Why the standard flat EV surcharge is wrong in two directions simultaneously, what the claims data actually shows, and how to build a severity model that handles the bimodal str...
Conformal prediction gives finite-sample valid 99.5% risk bounds for individual policies — useful for premium risk SCR validation and model validation consistent with Solvency I...
Li and Castro-Camilo (arXiv:2603.23309, March 2026) unify inverse probability weighting and extreme value extrapolation in a single estimating equation. Here is what it does, wh...
TabPFN v2 (Nature 637:319–326, 2025) does zero-shot prediction on datasets up to 10K rows. Here is what that actually means for the pricing segments where your current models ar...
Avanzi, Richman, and Wüthrich reformulate individual claims reserving as a Markov Decision Process. We explain why it matters, what it actually does, and when a UK reserving act...
Chevalier & Côté (EAJ 2025) benchmark nine GBM variants on five insurance datasets. We read it so you don't have to, then show where insurance-distributional fits in.
Laub, Pho and Wong's ANAM paper enforces smoothness and monotonicity architecturally, not as penalties. Here is what the mechanism actually is, why it matters more than the benc...