I’m sorry! I should have made it clearer in the article, but there isn’t nearly enough information in the article — not even with the appendix — to reproduce the Monte Carlo analysis.
My target audiences are broadly those who commission models and those who build them — this is more directed to the former, hence the focus on mapping, getting buy-in around what is modelled (maybe up to the data used to condition the model) and then the results and how to interpret them. The actual modelling is backgrounded here, but I had to put in some detail or it wouldn’t makes sense!
If you’re interested I can send you the Excel / ModelRisk model I used to generate the results. It’s a bit belt and braces and, as it stands, it’s not very clear (no comments and built for speed rather than structural clarity).
Otherwise, I do plan to write a follow up article where I go through the modelling in some detail. This will be a while though. I have quite a few other things, I’d like to get out first and this is spare-time activity, so it easily gets knocked off the boil by the day job or domestic distractions!
In terms of introductory texts, that depends very much on your background and your ambition. This is an article in itself, but briefly:
If you’re comfortable doing a bit of math (high school level) then I really like Fenton & Neil’s Risk Assessment and Decision Analysis with Bayesian Networks (BN). It seems a bit arbitrary to focus so quickly on BNs, but actually they use them more like I use causal maps and they introduce all the math you need clearly and pedagogically more or less from scratch.
Bedford & Cooke’s Probabilistic Risk Analysis is broader and a bit more thorough, and covers most of the practical stuff you need (things like expert opinion, decision theory, project risk, regulation, etc). But it’s quite a bit more demanding in terms of pre-requisites. In theory it also starts from scratch, but you need to be fairly comfortable with university level math (economic, engineering degree level; it’s not THAT brutal).
My absolute stand out favourite book on these subjects is Tom Körner’s Naive Decision Theory. Tom writes, in theory, for high school students, but Tom is a professor of Pure Mathematics at Cambridge University, so his idea of the level of high school students is a bit twisted. It’s demanding, but so much worth the effort.
Less mathematical, Pearl’s Book of Why is very good on causal mapping and Doug Hubbard’s books are entertaining, if not massively helpful (very good on saying what everyone does wrong, but despite a valiant effort, not so great on how to do things better, which, to be fair, isn’t simple).
As for online — Alexei Sidorenko’s blog is a good place to start and there’s quite a bit of good stuff at Risk Awareness Week 2019 and 2020. Courses can be tricky because it’s hard to know exactly whether the content will help you with your particular concerns and it’s hard to know where the mathematical level will fall. That said, Doug Hubbard’s Applied Economics course is supposed to be very good and Strategic Decisions Group has some good (if rather pricey) course offerings.
If you’re sitting with specific issues, where you’re thinking some kind of modelling might help then I’d be happy to spend an hour (or so) to discuss it. No charge and no commitment. I’m not even that concerned that there is line of sight to potential work. There’s an immense amount of value to me in hearing what people are working with and I’m always happy to help where I can.