231 articles across 19 categories. The counts reflect actual post frontmatter — a post tagged in multiple categories appears in each relevant archive.
Core disciplines
Category
Pricing
Rate adequacy, burning cost models, credibility weighting, GLM structure, and the decisions that determine whether a book makes money.
123 articles →
Category
Techniques
Statistical and ML methods evaluated on whether they actually deliver in a UK pricing context. Not a textbook — a field report.
81 articles →
Category
Libraries
Coverage of specific Python libraries: what they do, how they work, and when to reach for them versus writing something yourself.
98 articles →
Category
Tutorials
Step-by-step worked examples with real code. Designed to run end-to-end. Motor pricing, credibility weighting, factor tables, GLM diagnostics.
45 articles →
Category
GLM
GLM structure, offset terms, interaction handling, and the gap between what Emblem does and what the underlying statistics says.
5 articles →
Category
Severity
Large loss loading, spliced distributions, extreme value theory, and the frequency-severity independence assumption most GLMs get wrong.
6 articles →
Category
Insurance Pricing
The broader craft of pricing — rate review process, technical price construction, multi-factor interaction, and the strategic questions above the modelling.
5 articles →
Advanced methods
Category
Causal Inference
DML, causal forests, TMLE, and regression discontinuity applied to pricing. When correlation isn't enough and you need to know what actually drives loss.
12 articles →
Category
Machine Learning
GBMs, neural networks, embeddings, and foundation models applied to pricing. With honest assessments of whether the complexity pays off.
6 articles →
Category
Research
Academic papers and research directions relevant to UK insurance pricing, with field notes on whether the methods survive real-world data.
6 articles →
Governance & risk
Category
Regulation
FCA Consumer Duty, PRA SS1/23, proxy discrimination, and model risk governance. Regulatory requirements translated into audit-ready Python workflows.
21 articles →
Category
Fairness
Proxy discrimination, protected characteristics, indirect use detection, and long-run fairness under feedback loops in live pricing.
12 articles →
Category
Compliance
Model documentation, audit trails, and explainability requirements. Python tools that produce the evidence files regulators ask for.
9 articles →
Category
Model Risk
MRM under PRA SS1/23. Risk tiering, inventory management, monitoring obligations, and how to defend a model under challenge.
8 articles →
Category
Model Validation
Independent validation — the process, required evidence, and tests that catch problems GLM diagnostics miss. Designed around PRA SS1/23 expectations.
5 articles →
Operations & quality
Category
Validation
Gini, double-lift, calibration, out-of-time testing, and the failure modes standard ML metrics miss on insurance data.
11 articles →
Category
Monitoring
Model drift, champion/challenger testing, PSI, A/E ratios, and covariate shift detection in live pricing. The discipline most teams under-invest in.
12 articles →
Category
Comparisons
Head-to-head evaluations: insurance-specific tools versus generic ML libraries, Python versus R, different approaches on the same dataset.
7 articles →
Category
Python
Python-specific tooling, package selection, Databricks integration, and running Python in a regulated pricing environment.
8 articles →