I think there is a fatal problem here apart from cost and complexity. A fatal problem that makes this device potentially DEADLY. Sorry for shouting.
Consider air. Add humans. Humans convert O2 and water and food into CO2 and H2O. Your machine takes out the CO2.
What happens is that the O2 levels will inexorably fall, until you basically have pure-ish Nitrogen. If a human or other mammal breathes pure Nitrogen they will not feel suffocation. Suffocation detection is triggered by CO2 levels which your device removes. What happens is that you would feel OK and then suddenly faint and subsequently die.
Here is a link to a euthanasia device that implements this mechanism as a way to euthanize yourself.
Vitamin D is out of patent so profit margins are limited. Same issue with hydroxychloroquine.
The study is fairly small, so the reduction in deaths (2/50 to 0/25) was not statistically significant. The dramatic reduction in ICU admissions was s/s though. From my perspective the room for doubt on the benefits of vitamin D3 is now very small since this study which was an RCT (randomized trial). I will certainly look at any large RCT if/when it comes in, I am in no state of suspense about this.
Apart from the studies mentioned above, there are numerous other indirect lines of evidence, e.g.:
1. Severity of the disease in migrant communities with dark skin or social mandates of covered skin e.g. Somalians in Sweden, African Americans in the USA. which inhibits D3 production.
2. Death rates in countries with high incidence of vitamin D deficiency e.g. Belgium, Italy, versus those with low levels (Scandinavia even Sweden, who eat oily fish and supplement/fortify).
3. Low impact in countries and communities (e.g. homeless people) with high sun exposure.
There are also very realistic mechanisms and explanations for why and how vitamin D3 would have this effect, and prior studies on the impact of vitamin D3 on respiratory tract infections including other pneumonias.
When looking at the literature in this space, note (as in virtually all areas of medicine) that bad studies abound. Some things to look for: excessively small studies seemingly designed to produce a not s/s result combined with the belief that a non-s/s result == proof there is no result; large intermittent bolus doses used that generate surfeit of D3 then a deficit; excessively small doses; failure to take the D3 with fat to ensure digestion; failure to take vitamin K2 with D3 for optimal results; failure to take into account accumulated deficits and obesity whereby many months or even years of D3 vanish without trace into fat stores; funding sources with vested interests in a certain outcome (e.g. osteoporosis medication suppliers with an interest in a finding that D3 is not useful in treating osteoporosis and you should use their far more expensive product) ...
Important points to add
1. Make sure you have enough data that the result is not due to chance. 10 years of monthly data is laughably insufficient IMHO.
2. Make sure that you did not let your subconscious brain conveniently cherry pick a place and a time of high performance as your backtest playground. That is, make sure the data is representative of at least the breadth of past scenarios. Protip: it is quite unusual for stock markets have such high returns. They rarely last. Have a look at the Japanese Stock market over the last 50 years. Or the Chinese stock market in recent years. Or the 1930s in the US.
3. Make sure your optimal leverage is not right next to a 'cliff' in the parameter space where even slightly more leverage would have wiped you out.
I did a similar analysis going back to 1926 for the US alone and found that the optimal leverage was a very fragile 10%. 20% did far worse.
Many people tell me they are bullish on America and so the USA outperformance (best in the world over many periods in living memory) will continue. OK, you are bullish on a country that will pick either Donald Trump or Joe Biden as the president for the next 4 years. Good luck!! Would you have been bullish on the USA when its outperformance started, shortly after a ruinous civil war, with rampant corruption and civil disturbances, only barely emerging from developing country status?
I would add that your narrative that this is a myth propagated by nostalgic macho cowboys is not evidence.
This is a lazy post. You can easily find papers on this in google scholar.
See e.g. doi: 10.1210/jc.2006-1375
"A Population-Level Decline in Serum Testosterone Levels in American Men"
There has also been a large decline in sperm counts. The declines seem to have started about mid C20.
At one point, sceptics were quite vocal in their view this was not real but they have gone quiet of late e.g. Professor David Handelsman.
It is interesting that because "normal" levels are based on population samples, the normal levels have been reduced in some places as a result. Levels considered normal now would have been considered seriously low not too long ago.
There is little or no interest in finding out why this is happening. Theories abound.
Estrogen mimics in foods (soy. industrial milk, grains and grain and soy based oils), pesticides, water supplies (excreted estrogen from women taking birth control). My own suspicion is that it is due to a combination of factors
Note in this space be aware that many of the studies have been industry funded and seemingly rigged to produce a desired outcome that product X "has no [statistically] significant effect." Which result is not surprising given that the study was seemingly made so small and of such a short duration that only a huge effect would give the magic p<0.05.
Great post. The political process selects for ability to construct narratives, tell stories, blame others for failures and take the credit for successes, etc. So that's what we get. Sometimes by chance we get more, very often not. The pandemic has laid bare the incompetence of many world leaders on both sides of the political divide.
It is also a great illustration of the power of politics in filtering people's reality.
Specifically on innumeracy we see again and again the rejection of simple measures that would reduce infectivity because they are not perfect or not high status. I am referring to some of the measures taken by Taiwan - with no lockdown and 1/1500 th the death rate per capita of the USA - such as universal use of face masks and screening people entering shops for symptoms and fever. These measures filter down the infectivity of the population as a whole such that r0(eff)<1 or close to that level, at relatively modest cost. In this podcast we hear from a doctor that masks and screening are only useful as ways of showing that we care. Really*. https://www.abc.net.au/radio/programs/coronacast/have-we-been-too-easy-on-rule-breakers/12524256
Drastic measures like lockdowns seem to be perversely popular because they seem to signal 'strong' leadership. Even though they are not very effective, part because you cannot lock down the whole society, and are enormously expensive. There is great faith in contact tracing, which simulations show only helps if it is very rapid and if test results come back fast and if quarantine is very strictly enforced. In many countries none of the above apply.
*Another filtering problem: The medical training system in large part filters for the ability to memorize vast amounts of material and for physical and mental stamina. And not much else. Sometimes by chance people get through this system who are statistically, mathematically and numerically literate but many get through who are not. Even researchers - as a perusal of the medical research literature quickly attests. Did you know that p>0.05 shows that there is no effect? People who took Vioxx and had a heart attack, more than 50,000 excess heart attacks in all, may disagree.
The US does pretty badly in the world tables for school performance in math especially considering its GDp/capita. https://www.oecd.org/pisa/PISA-results_ENGLISH.png
(Myth 5) Online learning: Online learning and online convex optimization
Can you be more specific about which result(s) within the 84 pages of this document substantiate your claims? Actually it would be better to have a reference to a result in a paper in a peer reviewed journal.
Interesting that, very often, people interpret a frequentist result as though it were Bayesian. E.g. that there is a 90% chance the true value is within the confidence interval. This is so common in medical research that it may possibly be the majority interpretation.