wasmania

After you finish wiping your arse with disinfectant and cleaning your kitchen counters with toilet paper, could you respond to Dynato's questions?

Epidemiology isnt in the 'how' business. Thats what medical doctors do. Epidemiology is about the mechanics of disease spread and how best to mitigate the spread.

understood. that's why I recommend you. you would present no vetting issues re: prior knowledge. An empty vessel to be filled, as it were.

I used to much science in my explanation - I said the the early models are missing dimensions and I gave an example of the lack of knowledge in early models about the impact of asymptomatic people spreading the virus that caused the less dense areas to be over forecast, I said that the modelers are modeling chaos, that they are objective but wrong. that they improve as the data improves but a true post mortem takes a long time - how many different F$%cking ways do you need it told to you.

I think there might be a future for you in the current administration

I have basically been agreeing with you on the current model quality. might you do me the honor of explaining what your motivations are to be such a blowhard against the epidemiology process?

If they are anywhere near as smart as the armchair blowhards on this site, I think they would be lined up at the high school auditorium tighter than sardines in a can.

Pretty sure those pork plant workers were pretty tightly situated so there is something of an apples to apples comparison. And yes, I know the Smithfield plant is in South Dakota not North Dakota, because that makes a big difference...

I have been trying to explain to you and others that models vary widely as data changes rapidly, which it has been doing. models can be both objective (built using statistical theory and proper data management) and wrong (missing key data points or dimensions that have not been capture or are unavailable at the time the model was developed). It is only way after the fact in a post mortem of the pandemic and full examination of the data over the entire time period can you fully understand what drove the pandemic and why. you want to kill these messengers who are trying to model chaos for reasons I still don't grasp.

ok lets just put you in the bucket of people not looking for the truth, only a win for your 'team'.

there are partisan slants unfortunately, some that are partisan and accurate and others that are bullsh$t. But as individuals with a common purpose (that is what we are despite the division sown by politicians and media) we should be trying to get past the rhetoric to get to the truth, rather than be pushed by our favorite pundit into irrational and harmful fighting amongst ourselves.

this may end up being true. So a geopolitical conspiracy to hide this unfortunate event might be happening. That is different from a partisan conspiracy to inflate the danger to make you lose your job and the president lose the election.

sorry, he made it a model when he took the current death rate and multiplied by 2 with the assumption that the U.S. is half way through.

the above is laughable. his argument centers around a graph of the current U.S. covid deaths to date vs U.S. and global annual flu deaths. His prediction from this (April 10th) is 33000 deaths in U.S. a number we are already passed. So this what you get from an anecdotal forecast by a partisan blogger. Egregiously wrong in the other direction. Perhaps the willingness to accept malice as a reason for scientific error is because you are so used to it from the unscientific partisan ranters that take up space in your head. 'If my guys do it, everyone must!'

stupid is the word used in Hanlon's law, I suggested ignorant as more accurate in this case. When the only data you have comes from early spread in high density areas, it wont do well predicting low density outcomes if the model is data starved. It was learned very late that asymptomatic people were big carriers. If this was known earlier, than a variable like % of people who daily use public transportation would probably be huge in the model and result in low forecasts for places like North Dakota. Not malicious, ignorant of a key dimension.

See my previous posts agreeing and pointing out that models driven by unknown dynamic phenemona are usually very inaccurate until the dynamism and other unknowns are captured in the data. So this is not a point of contention between us. My question is why cannot this be accepted as truth rather than a rush to say someone is cooking the books for malignant purposes? A version of Hanlon's law below: "Never attribute to malice that which is adequately explained by stupidity" (Stupidity meaning the ignorance inherent in early models that lack sufficient data history)

therefore what.... conspiracy?

This one suggests way more people are infected than what has been identified through testing. Looks more like a sober explanation but in a scientific lingo many people will be frustrated by.

I dont understand (rationally) why so many here are desperate to conclude that epidemiological models are a tool in a nefarious plot to keep them unemployed, cause a political downfall of their favorite candidate etc. Bad forecasts are a hallmark of modeling things that are poorly understood or incredibly dynamic. As the data improves the accuracy gets better. I urge you to give equal time to sober explanations of the forecasting process (Johns Hopkins, CDC etc.) , in addition I guess to the blogs and pundits that tell you what you WANT to hear.

Dang it. owned again.

I do understand that hypocrisy is a feature of message board addict chatter, not a bug. So when I point it out, I do expect a disengenuous 'literal' interpretation of my post. Thanks for not disappointing me.

But it doesn't matter because they were in a bad relationship before this happened, kind of like having an underlying condition so the covid doesn't count.

the point you make is redundant to many previous, but gotta hand it to you, you made it unique by adding the racism. great job.

Ok. I am a data science guy. MBA Stats. 30 years of logistic regression etc for commerce and credit. Models are only as good as the underlying data. How relevant is the past history? how should the current data be sliced by region/age/whatever etc. Many more data hygiene and timing questions. Any honest pro is using models to guide planning and knows full well that ' no plan outlasts contact with the enemy'. You seem to hold them to a higher stand for some reason.

because anybody can do epidemiology.