sorry... extended dog walk and then grilling dinner was in order. the pleasures of covid.Typical Lax Dad wrote: ↑Tue Sep 22, 2020 6:20 pmWho is IHME marketing their model to and how are they compensated....as for mask, a lot of the variability in their model is with/without masks. Are you in favor of no modeling at all? ( BTW, you brought up Masks before I did. It seems you were questioning the need for them in some places).wgdsr wrote: ↑Tue Sep 22, 2020 6:11 pmi'd imagine they're doing both.Typical Lax Dad wrote: ↑Tue Sep 22, 2020 6:06 pmHaving for what seems like a lifetime running and analyzing models, I take them with a grain of salt. I hope nobody in government assumes that the information has a lead pipe cinch confidence factor. Is that group marketing or publishing it’s data?wgdsr wrote: ↑Tue Sep 22, 2020 5:56 pmthanks for those. what it all means to me is... garbage in (novel virus, uncertainty cubed, limitations, bad and incomplete and faulty inputs) results in garbage out. so not just useless (in my opinion), but at least potentially damaging if marketed. as ihme clearly has been.Typical Lax Dad wrote: ↑Tue Sep 22, 2020 5:29 pmYou mean like this: https://www.acpjournals.org/doi/10.7326/M20-1565wgdsr wrote: ↑Tue Sep 22, 2020 5:23 pmtheir encapsulation now finally incorporates at least the words some prior coronavirus immunity, nothing about t cells or potentially faulty seroprevalence data although it's heavily relied upon... references social media... says most testing is in symptomatic individuals... i could go on --- but let's just say their analysis, conclusions, extrapolations look alarmingly simplistic and elementary, not to mention subjective and poorly sourced.Typical Lax Dad wrote: ↑Tue Sep 22, 2020 4:39 pmTrust was the wrong word. “Rely” is better word and as for directionality, there is a reasonableness factor.wgdsr wrote: ↑Tue Sep 22, 2020 3:32 pmof course you can trust them, they get more pub than anyone else.Typical Lax Dad wrote: ↑Tue Sep 22, 2020 3:20 pmTheir “model” isn’t “accurate” so you can’t trust them. I have yet to see an “accurate” LBO model but Private Equity continues to feast. Sometimes living with directionally accurate is good enough.DocBarrister wrote: ↑Tue Sep 22, 2020 3:13 pm It didn’t have to be nearly this bad. There is no way on Earth that a virus should ever be the second leading cause of death in the United States.
Covid-19 is now the second-leading cause of death in the US, just after heart disease, according to the University of Washington's Institute for Health Metrics and Evaluation (IHME).
What happens next with the pandemic largely depends on personal responsibility and how much Americans are willing to fight this battle together.
Already, Covid-19 has killed more people in the US than Americans killed in battle during the five most recent wars combined: the Korean War, the Vietnam War, the Iraq War, the War in Afghanistan and the Persian Gulf War.
https://www.cnn.com/2020/09/22/health/u ... index.html
I know many support Trump because he’s an ignorant, racist, bigoted, misogynistic piece of trash like them. But is one’s ignorance, racism, bigotry, and misogyny worth the deaths of so many people?
DocBarrister
can deaths go down directionally?
as it is, covid-19 has killed more americans than the last 5 most recent wars combined.
sourcing isn't easy, but... for me... they don't take a lot of the science and epidemiology that has transpired out there to date as much of a guide. on this latest "prediction", i will admit i haven't been able to figure out how they come up with this huge end-of-year spike, so shooting it down makes that more difficult. so i default to how wrong they've been to date on the modeling i've seen them do and what it looks like they are considering and drawing conclusions from. whatever it is they don't like in november and december, be good if they're wrong.
https://www.nejm.org/doi/full/10.1056/NEJMp2016822
Does prediction = model output?
here's from the group that they rely on, it looks like heavily, to judge mask wearing across the country... from a daily facebook survey. which they market heavily to their numbers depression. it's a world map from their survey on % of people with covid-like symptoms.
https://covidmap.umd.edu/map/
does that look accurate overall?
does it matter who is doing the marketing?
if you bought this group's lbo model, grain of salt or not you'd fire them. in my opinion!
i dunno. i get their marketing (or whomever is doing it for them). so do you. so does or did the prez. i don't know who's doing it. ihme? main street media? still the white house? before you'd posted your cnn headline, i'd just a bit before seen "200,000 deaths, just the tip of the iceberg" out of cnbc. granted, the headline itself wasn't from an ihme individual, but they led with one of the ihme guys and "the worst is yet to come". i don't know of course how they're compensated. and if there's no agenda, i don't rightly care.
modeling makes the world go 'round. it gets deals done. it's only as good as the quality of the modeling. i'm not in favor of bad modeling.
as far as masks... i'm fairly sure any referencing of masks in this discussion by me was related to how ihme was utilizing mask stats and inputs. including referencing that a great priority (variability) is them saying masks/no masks. don't think i questioned the need for them in this thread... but quote me and i'll try to detail?
my other issue just on the mask front above is... they have garbage in (facebook survey). and then what's the analysis and how they come by it for the change in results even if the data was good?
i fully appreciate that i don't have access to any supercomputer and phenomenal epidemiological construct that doesn't show out of their website and their quotes. so i'm left with their performance. do i have to be in favor of how they've come front and center into public consciousness and sometimes policy? i don't! anyway, model away!!! no way i'm stopping them!!!
speaking of masks -- what's with figure 9b !?!?
http://www.healthdata.org/sites/default ... 091820.pdf
i thought the deep blue states were supposed to be carrying this country with masks and the maga guys and gals were ruining everything?!?!?
and seriously, what's up with new england in particular up there? card carrying and everything. is ihme data accurate?