Follow up: UK data now showing new auto registrations down 32% cf pre pandemic YTD (> the 2020 GDP shock), except now order books (demand) are still STRONG. Economic models are unlikely to pick up this production input criticalness forecourttrader.co.uk/latest-news/ma… #SupplyChain

🧵 Modeling the distributional #epidemic-#economic effects of the #COVID19 #pandemic by Marco Pangallo Follow our thread for highlights! Streaming now: youtube.com/watch?v=z3L0ER…

Thanks Santa Fe Institute for hosting and for this nice thread with the highlights!

Next, we have data-driven ABMs in macroeconomics: a lot of data on households, firms and banks that get used to study effects of policies across the income distribution. See mdpi.com/1099-4300/22/2… & monetaristinheels.com, oecd-ilibrary.org/economics/the-…, sciencedirect.com/science/articl…

Excited that our paper on modeling the economic impact of the first wave of the Covid-19 pandemic on the UK economy is finally published! ✌️ Check out this excellent thread by @AnToniPichler to get an overview of our results 👇

📢 A paper that I really like just came out as a Bank of Hungary working paper: mnb.hu/letoltes/mnb-w… Let me explain why I think this is a great example of the usefulness of data-driven agent-based models in economics. 👇 1/7

This is super interesting! I will try and do a Twitter thread but in meantime I have written something about this and a live case study I have been working on linkedin.com/posts/robert-h…

This thread is the father of all straw-men justifications of (mainstream? neoclassical?) economics. All models are approximations, but why should we use e.g. rational expectations as a baseline and treat deviations from that as the exception? This is a precise choice. 1/4

Can the intrinsic physics of multicomponent systems show neural network like computation? Out now: nature.com/articles/s4158… Led by Constantine Evans w/ Jackson O'Brien, Erik Winfree. U Chicago summary: news.uchicago.edu/story/uchicago…

My paper writing process is a rather less efficient version of this. 1) Write a paper 2) Retitle it "Online Appendix" 3) Write a 2nd paper summarising the 1st 4) Retitle the 1st paper "Supplemental Appendix" & the 2nd "Online Appendix" 5) Write a 3rd paper summarising the 2nd.

Geoffrey Hinton says the more we understand how AI and the brain actually work, the less human thinking looks like logic. We're not reasoning machines, he says. We're analogy machines. We think by resonance, not deduction. “We're much less rational than we thought.”