One piece of evidence in favour of this is that the Scottish schools finish earlier than the English schools. Correspondingly, the Scottish case rates started dropping earlier than the English rates.
However the Northern Ireland dates don’t appear to match up so nicely—they have an early summer holiday but the case rates peaked roughly when England’s did.
So I’m not really sure!
Another consideration on the ratio between infections of unvaccinated vs vaccinated is the extent to which the different groups react to the current prevalence of Covid.
If vaccinated people relax more when prevalence is low but are much more cautious when prevalence is high then that would tend to make the ratio increase as witnessed.
This would make sense to me—being unvaccinated will be correlated with generally being less concerned about covid
Agreed.
I expect that resistance to long covid is somewhere between resistance to infection and resistance to hospitalisation. This might roughly double the expected numbers of deaths / long covid.
Even if all the Georgetown undergraduates were somehow infected by Delta
and
mRNA reduces these both by .9.
mRNA vaccines don’t decrease hospitalisation by .9 given someone has become infected, they decrease it by .9 compared to an unvaccinated person given typical community exposure. So I think your calculation is more like “Even if all the Georgetown undergraduates were exposed to Delta in a way which would be sufficient to infect them if they were unvaccinated”.
I would estimate that to get back to your original scenario we probably have to multiply by 3-5 (depending on how much of the resistance to hospitalisation you think is purely resistance to getting infected in the first place).
Is there a story behind the PC results 2008-2011 where they deviate from the fairly straight line?
Nitpick: There’s an issue with Magnus Carlsen’s ratings—he became a GM at 13 in 2003, not 21 in 2011 or in 2009 as shown in the graph (famously he was the second youngest ever GM at the time).
I don’t know if all incoherent creatures value everything the same amount—it seems like it should maybe be possible to just have a restricted region of incoherence in some cases, but I haven’t thought this through.
This puts me in mind of sacred vs secular values—one can be coherent and the other incoherent as long as you refuse to trade between the 2. Or, more properly, as long as only 1-way trade is allowed which seems like the way sacred vs secular values work.
That’s super-helpful—thanks!
One thing the negative meta-study noted was the variation in doses between studies (12 − 210mg).
Very interesting.
Is it possible to compare the transportation and/or assembly costs to the manufacturing costs? And in fact to compare these to the costs of manual threshing? I think this would give good context to the scale of the problems.
It seems like there would be a business to buy a machine and take it round various farms in the local area (assuming assembly really is easy-ish).
Yes, I think this definitely had some effect (I misread 10th May as 10th March originally!).
I also think the fundamental search must have been better in #1. Of the 13 present in #1 but missing in #2 6 were from 2020, 7 from 2021 (I haven’t looked at specific dates for the 2021 ones but I’d guess some of them are from before March 22nd).
For #1 the original search turned up 1237 articles then 95 full text analysed leading to the 19 included. Corresponding numbers for #2 are 256, 12 & 10.
I had a quick look and essentially it seems the latter found fewer studies (10 vs 19) and therefore fewer patients (1173 vs 2768)*
They have similar central estimates for RR of all cause mortality (0.37 vs 0.31) but due to having more patients the former has tighter CI (0.15 to 0.62) and concludes that there is an effect but the latter has wider CI (0.12 to 1.13) and concludes that there isn’t an effect.
The latter could claim that there is as yet insufficient evidence of an effect based on the studies in their analysis but not that these isn’t an effect. I especially take issue with the claim that “IVM is not a viable option for treating COVID-19 patients” when they themselves take such pains to talk about how low quality much of the evidence is!
The two meta-analyses also differ on their ratings of different papers—for instance the largest study (n=400, Lopez-Medina et al.) is rated as High quality (7 out of 7) in the former but at high risk of bias in the latter (due to deviations from intended interventions).
Scanning the paper there are a few issues. For the most part the problems are mitigated but there could still be issues:
There was a period of 17 days where the placebo group were receiving ivermectin(!)
The people from these 17 days were excluded from the primary analysis and additional subjects were enrolled
The primary outcome was changed during the study
This occurred ~30% through the study.
The reasoning for changing from the old outcome seems reasonable
It’s hard to comment on whether the selection of the new outcome could have seen bias
Placebo was changed during the study from dextrose water to something tasting more like ivermectin
This was ~25% through the study
Results did not differ much between the 2 placebo groups
I don’t feel like this would make a massive difference but I’m not sure
This paper is fairly typical of the quality of the studies (according to meta-analysis 2) or on the top end of study quality (according to meta-analysis 1) which causes me some concern.
In conclusion, if I was offered Ivermectin I would take it at this point (side effects seem to be small) and might even look to sign up to a trial if I had COVID—in the UK some people would be eligible for this one.
* 6 studies were common to both analyses.
Great collection of results. I particularly found the interactive graph useful.
I’m slightly confused by the trend lines (especially for Games and Other) - they don’t seem intuitively the best fits. It looks like they place a lot of importance on the high parameter recent models (possibly the cost for each datapoint is in parameter space rather than log(parameter) space?
My hypothesis—Apologising is low status.
Possibly this is no longer the case (apologising first is often seen as a sign of maturity) but I can certainly see this being the case in the ancestral environment.
This would match my experience that apologising is feels awful even when it is entirely my fault.
I guess if the other person has already apologised then me also apologising just puts our status back roughly where we started which is why going second feels much easier.
A previous calculation on LW gave 2.4 x 10^24 for AlphaStar (using values from the original alphastar blog post) which suggested that the trend was roughly on track.
The differences between the 2 calculations are (your values first):
Agents: 12 vs 600
Days: 44 vs 14
TPUs: 32 vs 16
Utilisation: 33% vs 50% (I think this is just estimated in the other calculation)
Do you have a reference for the values you use?
Or, alternatively, did Oxford really find a pharmaceutical company so incompetent that they did this by mistake, on top of giving an entire trial segment the wrong dose of vaccine the first time around? These are some rather epic screwups.
My experience working for a large company makes me not particularly surprised by this and I would give a decent amount of probability to this being an accident. I don’t know enough about the specific procedures to be hugely confident but it does seem most likely to me.
If we’re fairly confident that the wrong dose thing was an accident—I can’t think of any reason to do this deliberately and then try to cover it up—then AstraZeneca obviously have the potential to make big mistakes.
One scenario would be that the person requesting / approving the press release is not the same as the person running the project but rather their boss or their bosses boss or even in another department. The press release approver is less involved in the minutiae and has remembered the 79% figure, maybe even goes so far as to check their e-mails that this is the correct figure (or check with someone else who checks their e-mails). Probably none of these people were in the meeting with the safety board.
I have had this experience myself on many occasions where my superiors have given information to customers that is outdated just from them not being as up to date or forgetting the latest results. Obviously I’d like to think something like this would have more care taken about it but the dosing debacle is suggestive that checking things isn’t AstraZeneca’s strong suit.
That combined with the 0% chance of this not being noticed suggests to me that this wasn’t on purpose.
There is a system for describing human facial expressions—Facial Action Coding System.
This has also been expanded for some animals (chimps, macaques, gibbons, orangutans, dogs, cats, horses). Alas, no dolphins.
I wondered whether a decent amount of the cost increase was in changing from a hatchback to a sedan but I see that this is only $1,000 to go from the Mirage hatchback to sedan. And the Mirage sedan is the same size as a 90′s Ford Escort sedan/station wagon so size doesn’t explain it either.
Yeah, I didn’t actually answer q18 either (possibly knite maybe used my list as a basis?) for exactly that reason. Scott just put me in as the same as him for that question for the purposes of making an apples-to-apples comparison which seemed fine—no idea what I would have put if I had answered!
I’m kicking myself on #16 - I don’t know enough about epidemiology to make such a strong guess.
Yeah, I did a similar thing on #38 where I was similarly overconfident on an economy question which I don’t know nearly enough about.
On #16 itself I was lower than I should have been because I was using “virus” as a reference class rather than “respiratory virus” which was an obvious mistake looking back at it.
It seems Netherlands summer holidays depend on which region you’re in (North, Central, South) each of which is separated by a week (10th, 17th, 24th July respectively). However it seems that no matter which region you look at the peak is pretty close to the 17th.
Possibly with enough mixing between the regions this still kinda works but I feel like it’s a strike against the theory.