wtf is leeds_lectures_2026?

This repository contains teaching material for a two day training seminar on economic modeling held at the University of Leeds in July 2026. It presents nine small macroeconomic models arranged in a three by three grid. Three classic stock flow consistent models, called SIM, PC and BMW, each come from earlier academic work, and each one is written in three different coding styles: an aggregate version, an input output version, and an agent based version. Reading across a row shows the same underlying economy represented at different levels of detail, from one economy wide aggregate, to a set of connected industries, to a population of individual interacting households. Reading down a column shows the financial side of the economy getting richer, moving from a model with just government money, to one that adds government bonds, to one that adds bank loans and deposits financing physical capital investment. Stock flow consistent models track every sector's balance sheet and every transaction between sectors so that nothing is created or destroyed by accident: every payment is recorded as someone's income and every financial asset is recorded as someone else's liability. The material explains this accounting logic before walking through each of the nine models in turn, and it also includes a companion book style version of the same content as a single readable HTML page. All nine core models run in plain R with no extra packages needed. A handful of optional packages are needed only if you want to regenerate the animated charts, network diagrams and Sankey diagrams shown in the material itself, and the README lists exactly which packages go with which script. The material links to several related lecture series by the same author covering similar stock flow consistent modeling topics for other audiences and events. The content is shared under a Creative Commons non commercial license, with credit required for reuse.