6 ms·
>> Presumably their statistical model indicated they won't be able to contain the spread which is why the current stated aim is to delay the spread[0]. What st
by YeGoblynQueenne 7y ago
>> Presumably their statistical model indicated they won't be able to contain the spread which is why the current stated aim is to delay the spread[0].
What statistical model? They're talking of "careful modelling". In the article you link, they don't mention anything about statistics, or, say, a simulation etc. You're assuming too much when you're assuming a statistical model, or anything based on data, or any concrete sort of model at all. "Modelling" can just mean a bunch of officials sitting around having tea and brainstorming potential scenarios.
- GrantZvolsky 7y ago> "Modelling" can just mean a bunch of officials sitting around having tea and brainstorming potential scenarios. Good point, but it doesn't seem to be the case. From the Coronavirus action plan[0] published on March 3 2020 > The UK is a world leader in the field of outbreak modelling and data analytics. The NIHR HPRU in Modelling Methodology led by Imperial College London has developed novel analytical and computational tools which exploit novel data streams on infectious diseases such as COVID-19. This group and other leading academic groups have developed tools to prepare for infectious disease outbreaks, which include real time infectious disease models, allowing policy decisions to be made using the best possible data and are actively modelling questions of relevance to dealing with the COVID–19 outbreak. [0]: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/869827/Coronavirus_action_plan_-_a_guide_to_what_you_can_expect_across_the_UK.pdf https://assets.publishing.service.gov.uk/government/uploads/...
- YeGoblynQueenne 7y agoWell, Imperial is a fine research institution indeed, but this still doesn't cut the mustard. Sure, Imperial boffins may be creating "real time infectious disease models" and those models may "allow" policy decisions to be made, but this tells us nothing about what we really want to know: whether these models were actually used to take the decisions we're discussing, instead of some other process. Really, the claims of having done "careful modelling" absent any evidence of that modelling (say, a graph or a table of the results) is a really bad sign.