Fine, I'll run a regression analysis but it won't make you happy • Hacker News Board

Re: Fine, I'll run a regression analysis but it won't make you happy

#231

Post by firesteelrain » Mon, Oct 02, 2023, 2:34 AM UTC

slv77 wrote: post #216

gpt5 wrote: post #201

Earlier quoted context omitted.

Initial constraints like ICU bed shortages and gear scarcity definitely had a basis. But those measure lasted long after the bottlenecks got resolved and the risk assessment became clearer. For example, public schools stayed remote while virtually every private school switched to in-person (even California's governor opted for private in-person schooling for their kids). The public's frustration is that these prolong…

ICU capacity was a bottleneck that never got resolved. In the US there was never a nationwide effort to mobilize ICU resources to target hot spots other than an initial aborted attempt to gather respirators for the surge in NYC and send a military floating hospital. Regional programs were put into place to shift patients during a local surge and to mobilize hallways and other non-traditional capacity. In California o…

This has been studied ALOT lately. Here is one such paper

https://ieeexplore.ieee.org/document/9383939

Re: Fine, I'll run a regression analysis but it won't make you happy

#232

Post by GenerocUsername » Mon, Oct 02, 2023, 2:42 AM UTC

slv77 wrote: post #172

Waterluvian wrote: post #131

It was a hard thing to broach at the time without people frothing at the mouth about how I must be anti vaccine or something: Has there been any attempt to calculate the cost-benefit of all the various measures and how extreme we should go with them? I’m guessing it’s hard to quantify and compare. A lot of things like general depression, isolation, kids missing half a year of school, etc. can’t really be evaluated ag…

The cost-benefit calculations in terms of excess deaths are irrelevant. It was clear even during the initial outbreak that policymakers primary constraints were the due to uncertainty about the virus and limits on availability of protective gear and ICU beds. Without protective gear there was no way to reduce transmission without isolation and once ICU beds are at capacity mortality rates double and people are dying…

[flagged]

Re: Fine, I'll run a regression analysis but it won't make you happy

#233

Post by GenerocUsername » Mon, Oct 02, 2023, 2:44 AM UTC

slv77 wrote: post #216

gpt5 wrote: post #201

Earlier quoted context omitted.

Initial constraints like ICU bed shortages and gear scarcity definitely had a basis. But those measure lasted long after the bottlenecks got resolved and the risk assessment became clearer. For example, public schools stayed remote while virtually every private school switched to in-person (even California's governor opted for private in-person schooling for their kids). The public's frustration is that these prolong…

ICU capacity was a bottleneck that never got resolved. In the US there was never a nationwide effort to mobilize ICU resources to target hot spots other than an initial aborted attempt to gather respirators for the surge in NYC and send a military floating hospital. Regional programs were put into place to shift patients during a local surge and to mobilize hallways and other non-traditional capacity. In California o…

Didn't that huge hospital ship in New York sit unused until it eventually left

Re: Fine, I'll run a regression analysis but it won't make you happy

#234

Post by refulgentis » Mon, Oct 02, 2023, 2:53 AM UTC

ezekiel68 wrote: post #222

I was a good US citizen and used what little influence I had among my social circle to encourage others to follow the COVID protection measures. And I adore Nate Silver. So, pardon me if I confess that all I read in this article was: Blah blah blah 'I mostly agree with ... an October 2020 anti-lockdown statement signed by a large number of scientists and medical professionals. “Those who are not vulnerable should imm…

There's some people who are still manage to be newly wrong even today.

For example, asserting that the government was enforcing strict lockdown measures in October 2020, if ever. This is despite obvious evidence that people preferred staying home, such as the still historically low rate of people working from offices.

I wasn't wrong once, from the first videos from China in January 2020. it's been amusing to watch people overthink and moralize and become very rabid over side alley of side alleys of 3 simple facts a high schooler knows: "very contagious respiratory disease bad, vaxx's don't prevent infection or transmission, people like working from home"

Re: Fine, I'll run a regression analysis but it won't make you happy

#235

Post by slv77 » Mon, Oct 02, 2023, 2:55 AM UTC

GenerocUsername wrote: post #233

slv77 wrote: post #216

Earlier quoted context omitted.

ICU capacity was a bottleneck that never got resolved. In the US there was never a nationwide effort to mobilize ICU resources to target hot spots other than an initial aborted attempt to gather respirators for the surge in NYC and send a military floating hospital. Regional programs were put into place to shift patients during a local surge and to mobilize hallways and other non-traditional capacity. In California o…

Didn't that huge hospital ship in New York sit unused until it eventually left

The hospital ship was never intended to take covid patients. The idea was the ship would take non-covid cases leaving ICU capacity in hospitals. It turns out that military hospital ships aren’t great for the general population given bulkheads and the general layout of ships.

Re: Fine, I'll run a regression analysis but it won't make you happy

#236

Post by plaidfuji » Mon, Oct 02, 2023, 2:59 AM UTC

Although the body of this article is about Covid policies and vaccine usage, which seems to have set off a whole round of discussion here on that topic, my sense was that the point of this article was more general:

> So my aim is generally to focus on stylized facts that are true and robust. And to keep repeating them. I like simple (or simple-seeming) claims that — and I can’t emphasize this last part enough — I expect will hold up to scrutiny.

As somebody who presents results of analyses and models for a living, I couldn’t agree more. Present the simplest possible finding that gets the point across and move on. Maybe mention “and this holds up when accounting for x, y, z”, or just keep those analyses in the backup slides in case you have a feisty audience member. Nobody cares how complex your model is, just that it works.

Re: Fine, I'll run a regression analysis but it won't make you happy

#237

Post by Grimburger » Mon, Oct 02, 2023, 3:01 AM UTC

Waterluvian wrote: post #131

It was a hard thing to broach at the time without people frothing at the mouth about how I must be anti vaccine or something: Has there been any attempt to calculate the cost-benefit of all the various measures and how extreme we should go with them? I’m guessing it’s hard to quantify and compare. A lot of things like general depression, isolation, kids missing half a year of school, etc. can’t really be evaluated ag…

> Has there been any attempt to calculate the cost-benefit of all the various measures and how extreme we should go with them?

In Australia there is huge pushback politically about holding an inquiry to determine what could be done better, a lot of politicians hid behind faceless "health advice" for their actions that wouldn't last a minute when the public servants involved face an inquiry.

Current estimates of the jobkeeper program put it at $110,000 per job saved, in comparison to the median wage of $52,338. Basically all of it went into the pockets of big companies rather than paid directly to people. That was just one of many programs at the time, debt blew out by hundreds of billions which the populace on both sides strangely cheered on.

Re: Fine, I'll run a regression analysis but it won't make you happy

#238

Post by jonahhorowitz » Mon, Oct 02, 2023, 3:01 AM UTC

ezekiel68 wrote: post #222

I was a good US citizen and used what little influence I had among my social circle to encourage others to follow the COVID protection measures. And I adore Nate Silver. So, pardon me if I confess that all I read in this article was: Blah blah blah 'I mostly agree with ... an October 2020 anti-lockdown statement signed by a large number of scientists and medical professionals. “Those who are not vulnerable should imm…

The measures that seem like they should have worked, but people claim they didn't are:

- masks: clearly they worked for the medical professionals that were treating patients during the first waves, as they were directly exposed and didn't get sick, why didn't they work for the population at large? Did we not wear them correctly, consistently?

- schools: It seems obvious that schools could be a huge vector of transmission, kids largely were asymptomatic and when they got sick, they would still spread it, both inside the school and to their vulnerable relatives. Were the school shutdowns a mistake? Should they have been done differently?

Re: Fine, I'll run a regression analysis but it won't make you happy

#239

Post by lethoso » Mon, Oct 02, 2023, 3:02 AM UTC

pwarner wrote: post #228

ezekiel68 wrote: post #222

I was a good US citizen and used what little influence I had among my social circle to encourage others to follow the COVID protection measures. And I adore Nate Silver. So, pardon me if I confess that all I read in this article was: Blah blah blah 'I mostly agree with ... an October 2020 anti-lockdown statement signed by a large number of scientists and medical professionals. “Those who are not vulnerable should imm…

Almost everyone was pretty wrong about COVID, at least at some points in time. But it's very rare for anyone to admit it. I'll admit I was wrong, between sanitizing my hands when it was really in the air, to assuming the vaccines would end it. Lots of mistakes on my end!

Sanitizing your hands, that must really haunt you.

Re: Fine, I'll run a regression analysis but it won't make you happy

#240

Post by vharuck » Mon, Oct 02, 2023, 3:02 AM UTC

todd8 wrote: post #128

Here are a few observations. From the CDC [1], 2020 total covid deaths by state: West Virginia 1318, Florida 21546, Maine 344, Vermont 134. The corresponding April 2020 census populations [2] for these four states are: West Virginia 1793716, Florida 21538187, Maine 1362359, Vermont 643077. Thus, before the vaccines for covid were available the deaths per 100,000 persons for these four states are: West Virginia 73, Fl…

I'd say the pre-vax differences likely came from how people treated public health advice and precautions. If somebody believed it was no worse than the flu or even didn't exist (like some of my in-laws in Summer 2020), they would be far more likely to catch it, spread it, and receive poor treatment.

Those differences in behavior would carry over to vaccination rates, since that's also responding to public health recommendations. It's hard to believe Silver didn't do the analysis on pre-vax death rates, but he didn't mention it so I don't know.

If COPD and similar conditions were the major difference, I would expect the difference in death rates to be closer to 2.3x (13.6%/5.9%) than 2.9x (73/25 per 100k). Those are very different ratios. And I don't know about any other complicating conditions with larger relative rates than 2.3x for those two states. Obesity is only 1.3x higher in West Virginia. So this gets back to Silver's point: you can almost always find other contributors, but that doesn't necessarily prove something else didn't contribute. You have to do that analysis all together. But complicating health conditions don't seem to be able to account for the differences, even before a regression test.