[ABE-L] Fwd: {MEDSTATS} Re: Covid-19 modelling

Sex Maio 22 11:22:28 -03 2020

Enviado do meu iPhone

Início da mensagem encaminhada:

> De: Bruce Weaver <bweaver em lakeheadu.ca>
> Data: 22 de maio de 2020 09:47:20 BRT
> Para: MedStats <medstats em googlegroups.com>
> Assunto: {MEDSTATS} Re: Covid-19 modelling
> Responder A: medstats em googlegroups.com
> 
> Perhaps there is something to be learned from meteorologists, who rely very much on ensemble forecasts these days.  I've not read all of it yet, but this set of notes appears to give a nice overview:
> https://www.e-education.psu.edu/meteo3/node/2284
> 
> 
>> On Thursday, May 21, 2020 at 7:19:18 PM UTC-4, John Whittington wrote:
>> On Tuesday, May 19, 2020 at 9:33:49 AM UTC+1, william... em gmail.com wrote: 
>> >John,  Good morning, great to see that your "normal" e-mail is "back 
>> >in town", and thank you for this reply. 
>> >Though my IT programming knowledge is rudimentary, i.e. I learnt 
>> >BASIC on Sir Clive Sinclair's ZX80, ZX81, and Spectrum at prep 
>> >school in Thanet 40 odd years ago, I do have a "basic" (ho, ho, ho) 
>> >question re these models in academic circles in Blighty and indeed 
>> >globally: how many other unis or "cutting edge" research centres are 
>> >performing modelling "on a par or better" than Imperial College's 
>> >please? Or does the latter have an effective "monopoly" or near 
>> >monopoly position? Surely competition is mandatory for optimal 
>> >research, discoveries, continuous improvements, quality, and 
>> >results? At least that's how gold medals are won at Olympic Games... 
>> 
>> It is, indeed - but in the case of the Olympic Games, there are 
>> established (and very simple to apply) ways of determining who should 
>> be awarded a gold medal.  In the case of real-time modelling of a 
>> pandemic-in-progress, one cannot determine who is 'the winner' until 
>> 'after the event' (aka 'when it is too late'!). 
>> 
>> However, to partially answer your question, a Google search for 
>> "Covid-19" modelling (WITH the quotes) gets about 677,000 hits.  If 
>> one adds 'UK' to the search terms, one still gets tens of thousands 
>> of hits.  Whilst the majority of those hits are undoubtedly 
>> 'irrelevant', it is an indicator of how much modelling is going 
>> on.  Countless ('cutting edge') universities and relevant 
>> organisations, both in the UK and throughout the world have been 
>> involved in modelling - so the short answer to your question is that 
>> there is plenty of potential competition, and certainly no 'monopoly' 
>> - but, as above, there is a grave problem in determining (in 
>> real-time') who is 'the winner' (or anywhere near being the winner), 
>> or even the 'best buy'!  Both national governments (like the UK's) 
>> and globally (like the WHO) have expert advisory 'modelling groups' 
>> (e.g. a subgroup of SAGE) who draw their membership widely from 
>> throughout academia and relevant institutions/organisations. 
>> 
>> The modellers and modelling groups that one hears most about (like 
>> Ferguson/Imperial) are probably those who 'make the most noise'! 
>> 
>> Whilst I'm sure that a lot of the modelling can be criticised for one 
>> reason or another, as I said before I think it should be on the basis 
>> of the (implicit or explicit) assumptions/parameters, and perhaps the 
>> algorithms, not the programming language used.  By analogy, if I had 
>> the ability (which I certainly don't!), I could write a PhD thesis 
>> about, say, a scientific topic, in Latin - but that, in itself, would 
>> in no way mean that there was necessarily anything wrong with the 
>> content, even if a modern-day scientist did not understand the 
>> language in which it was written (but perhaps criticised the work for 
>> that reason)! 
>> 
>> Kind Regards, 
>> 
>> John 
>> 
>> ---------------------------------------------------------------- 
>> Dr John Whittington,       Voice:    +44 (0) 1296 730225 
>> Mediscience Services       Fax:      +44 (0) 1296 738893 
>> Twyford Manor, Twyford,    E-mail:   Jo... em mediscience.co.uk 
>> Buckingham  MK18 4EL, UK 
>> ---------------------------------------------------------------- 
>> 
> 
> -- 
> -- 
> To post a new thread to MedStats, send email to MedStats em googlegroups.com .
> MedStats' home page is http://groups.google.com/group/MedStats .
> Rules: http://groups.google.com/group/MedStats/web/medstats-rules
> 
> --- 
> You received this message because you are subscribed to the Google Groups "MedStats" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to medstats+unsubscribe em googlegroups.com.
> To view this discussion on the web, visit https://groups.google.com/d/msgid/medstats/81b6777b-e3ea-4e1f-b596-556ebcf94d60%40googlegroups.com.
-------------- Próxima Parte ----------
Um anexo em HTML foi limpo...
URL: <https://lists.ime.usp.br/archives/abe/attachments/20200522/57c7dca2/attachment-0001.html>