##### articles

This page contains an overview of all my articles published on Medium, broadly categorized under the themes below. Some of the articles belong to several themes, so they’re listed multiple times.

Follow the links in the theme list to see the articles.

##### A word on the mathematical level and the amount of explanation

The mathematical articles are aimed at readers with a reasonable facility with high-school (A-level / gymnasium) mathematics. More advanced concepts should be introduced and explained in the article or sometimes in an earlier article - please let me know if they aren’t.

I’ve always maintained I never really understood A-level mathematics till I started my Bachelors, and my Bachelors only made sense when I read for my Masters. The main reason I did a post-doc was to understand my PhD. As such, it’s likely that my mathematical target group is professionals with a numerate degree (e.g. engineering or economics, not necessarily math or physics) even if you haven’t used it for a while.

That said, all the articles are designed to be understood at some level, even if the mathematics is not always entirely accessible. In many articles, the mathematical detail is left to an appendix.

I have marked the mathematical articles as follows:

🦆 - should be accessible to anyone who isn’t completely math-phobic, with a little patience.

🦉 - probably requires some comfort with high school math or a bit more patience.

🦅 - requires functioning solid high school math. The more math you have, the less patience you will need.

If not so marked then the article requires no mathematical pre-requisites.

##### Theme list:

The logos of mathematical modelling

Reflections on the philosophy and practice of mathematical modelling

The techne of mathematical modelling

The craft of mathematical modelling, especially (but not exclusively) modelling uncertainty, decision analysis under uncertainty, probabilistic inference in general and causal inference in particular

Quantitative risk management and decision analysis

Reflections on the practice and craft of using quantitative models to build strategy, manage risk and guide decision processes.

The particular mathematical challenges of capturing and modelling uncertainty in the highly volatile world of oil and gas exploration

Reflections on how to model decisions in a pandemic and how to make decisions with pandemic models

The logos of mathematical modelling

The Democratization of Mathematics

The demystification of mathematical modelling is the first step to raising the level of mathematical literacy in the public debate.

Mastery of mathematics for accomplished non-mathematicians

How high functioning in other intellectual disciplines can be an obstacle to learning mathematics.

Why probability theory is hard

It's not because you're stupid or weren't concentrating in school

The two schools of probability theory 🦉

A layman's look at the foundations of frequentism and Bayesianism and how you can have the best of both.

The problem with the prevailing paradigm of data analytics

The prevailing paradigm of data analytics is rooted in the philosophy of logical positivism. Institutions who do not primarily deal with data need a more flexible approach

The myths of modelling: Data speak

Data keep us honest, but they don’t speak; they aren’t objective, and they are never free from the taint of theory

The myths of modelling: Falsification

We do not verify models by repeated attempts to falsify them, nor should we try. With causality and Bayesian probability, we can do better

What does it mean that a probability is "correct" and how could you possibly know?

There’s more to a model than the fidelity of its forecast

“You don’t know what you don’t know” given a patina of academic respectability

The difference between vicious and virtuous simplification

In defence of armchair epidemiology

Armchair epidemiologists, the lackadaisical luminaries of LinkedIn; also have a role to play.

The techne of mathematical modelling

Transforming Scores Into Probability 🦅

How to turn consistently harvested assessments (scores) into probabilities

Method for seeing how good you are at predicting probabilities, probably

An elegant and insightful reformulation of Bayes’ theorem due to E.T. Jaynes

A causal explanation for how to avoid goats

A toy model of a the most serious modelling challenge the world has ever faced

The perils of extreme percentiles 🦅

Never use P1 and P99. Ever.

Swanson’s mean isn’t very good and we can do much better.

Probability Audit (free Excel tool) 🦉

Free tools for scrutinizing the uncertainty in probabilistic predictions

Tools for assessing probabilistic prediction

What is wrong with simple COVID models

The simplest COVID model boiled right down to basics

Quantitative Risk Management and decision analysis

First in a series of three articles outlining a transformation from risk matrices to quantitative risk management

Emerging from the muddle of matrices 🦆

Second in a series of three articles outlining a transformation from risk matrices to quantitative risk management

The Redemption of Risk Management

Third in a series of three articles outlining a transformation from risk matrices to quantitative risk management. If you only have time to read one of them, read this one.

Quantitative risk management 🦉

Illustration of the construction of a simple stochastic model for risk management

The Monty Hall problem as an example in constructing a causal map for decision making

Criticisms of the early modelling of COVID are unfounded, but we could have done better on communicating the rationale for lockdown

The Ineluctable Logic of Lockdown 🦆

Simple arithmetic shows that herd immunity was never going to work

Top ten probabilistic pitfalls in oil and gas exploration 🦆

The common mathematical mistakes I meet when reviewing exploration workflows

Out of the pit: Alternatives to the top ten probabilistic pitfalls in oil and gas exploration 🦆

How to fix the common mistakes I meet when reviewing exploration workflows

The perils of extreme percentiles 🦉

Extreme percentiles are parameters you can’t know at a place they have no relevance

Swanson’s Swansong 🦉

Swanson’s mean isn’t very good and we can do much better.

Probability Audit (free Excel tool) 🦉

Free tools for scrutinizing the uncertainty in probabilistic predictions

Tools for assessing probabilistic prediction

Method for seeing how good you are at predicting probabilities, probably

COVID modelling

Criticisms of the early modelling of COVID are unfounded, but we could have done better on communicating the rationale for lockdown

A toy model of a the most serious modelling challenge the world has ever faced

The Ineluctable Logic of Lockdown 🦆

Simple arithmetic shows that herd immunity was never going to work

“You don’t know what you don’t know” given a patina of academic respectability

What is wrong with simple COVID models

The simplest COVID model boiled right down to basics

In defence of armchair epidemiology

Armchair epidemiologists, the lackadaisical luminaries of LinkedIn; also have a role to play.