We need to see what the flu deaths are this year, with social distancing and lockdowns in effect, before we compare this to the flu, unless we are to believe that if everything had stayed completely open, this many people would have died.
1 Like
Which would be an interesting study to work on. Unfortunately we do know that flu is seasonal and we were well out of the season before distancing and other measures were put into place, so sadly I don’t think it will work this year. But I would love to see the data from flu+distancing vs covid+distancing.
I agree. Unfortunately, they’re all we have at the moment until more data gets put out and vetted. And regardless of the ultimate accuracy, SOME numbers are always better than NO numbers or purely guestimated numbers.
Yes. First, it’s 1 study, not multiple studies. Second, I have it on my hard drive. Third, it was a statistical modeling study using a small sample pool and a unique local environment. We need more data and we need people to validate this before we can say it’s true.
Finally, the emphasis is on MIGHT BE. I’m not shitting on the study, because it’s valuable, but you can’t just take it and run with it because you want the conclusions to be true.
Do you have a background in research?
Once again, though I’ve posted this elsewhere:
2018-2019: 34,157
2017-2018: 61,099
2016-2017: 38,230
2015-2016: 22,705
2014-2015: 51,376
2013-2014: 37,930
2012-2013: 42,570
2011-2012: 12,447
2010-2011: 36,656
Average of the past 9 years that the CDC has on record (Estimated Flu-Related Illnesses, Medical visits, Hospitalizations, and Deaths in the United States — 2018–2019 Flu Season | CDC) puts us at 37,463 people per year dead from the flu.
March 20th-April 20th puts us at nearly 42,000 deaths. Meaning we surpassed the 9-year average per year, in a month.
And everyone seems to be comparing the covid total projected deaths for the year to the particularly bad year where we got ~60K flu deaths, except, as I’ve said a million times before, to the point that I’m beating my head through the wall, these are the projected deaths with all the extreme, unprecedented, unheard-of-in-our-lifetime measures that are already implemented. So, whether or not it miraculously TURNS OUT to be less of an impact than the flu, comparing it to the flu at this point is just silly, except to say that they’re both viruses, which is about as broad as you can get.
7 Likes
In medical research? No, I’m a long short equity portfolio manager.
Not like it’s just the left saying this.
I do not envy you your job in the current market!
The reason I asked is that one thing that you are used to from your job is looking at numbers or metrics that are almost completely defined. Market volatility and human motivation/actions are always unknowns to be dealt with, but the actual metrics can be quickly defined and bracketed–indexes, stops, put-call ratios, advance-decline, acceleration bands, oscillators, many more numerous formulas I haven’t a clue about, you name it.
You’re obviously a really smart guy. The difference is that the number of unknowns in the current pandemic is far wider, and thus all numbers much more suspect (including the Santa Clara study you mentioned). Fundamentally, the metrics aren’t even established yet–we haven’t even figured out the best approach to gathering the data to define the metrics.
What I’m trying to say is that research is inherently much more gray than financials (which are PLENTY gray) due to the fundamental differences in how numbers are established. For example, studying the basics of how long the virus can survive on surfaces is important for obvious reasons–and so far we have one study that says under 3 days, one that says up to a week, and one that indicated it could be as much as 2 weeks. Each study had different methodology and circumstances. There’s no agreement at all, but over time it will coalesce around a more reliable number as more work is vetted.
So the true extent of carriers or asymptomatic cases cannot be bracketed right now. Research unfortunately moves much more like a glacier–the immense bulk of the ice grinding down and smoothing everything–than say, a lightning strike or a rock slide.
I can’t grasp the complexities of your job so this analogy is likely worthless, but it might be similar the situation that would occur if nobody knew what the True Average Range for an instrument was, or agreed on how to define it.
The only way to work the problem in this research–and I’m a mechanistic, physical/structural guy not a biostatistician like one of our other posters–is to triangulate. Unfortunately neither the media nor most people like this approach!
Edited–typos
2 Likes
My job is great right now, actually, this has been the best year for active management since 2008 thus far. Being short a lot of garbage has led to a great YTD gain, so I’m hoping that continues.
I can fully appreciate that the unknowns for a pandemic are far more than looking at financial markets or individual equities. My point was, and still is, simply that these unknowns are not being reflected in the current figures. So when people are saying things like “oh the death rate of Covid19 is X%”… well, it’s simply not. People don’t need to extrapolate the Santa Clara study and assume the true mortality rate is 0.3% or less, but my point is it’s highly likely that it is substantially lower than the current reported figures.
We likely won’t know the real mortality rate for a long time, and even then the government has incentive to inflate it to justify their martial law, excuse me, “stay at home” experiment. Kinda hard for the government to come out and admit the death rate is 0.2% and that’s why they had people locked in their homes for months. So either way I think it’s unlikely we get reliable figures regardless, which is unfortunate.
2 Likes
Like @Aragorn said I’m not sure the timing works. However, I would be interested to see how the number of deaths for known underlying conditions is affected by this virus. I suspect we will see deaths from cardiovascular disease, cancer, and diabetes down in 2020.
2 Likes
We have a seasonal disease here, vinterkräksjukan (Calicivirus)
and we now have zero cases as a result of distancing and handwashing.
^
| number of cases per week
→ week
2 Likes
We’ll also see fewer diagnosed cases of blood cancer but it won’t mean that people aren’t still developing blood cancer.
Ya, I think people will continue to develop these issues. I was talking specifically about death since people with certain pre-existing conditions are dying more often from the coronavirus.
Apparently, researchers discovered the first death was actually February 6th not the end of February like previously thought.
https://www.wfsb.com/news/two-californians-died-of-coronavirus-weeks-before-previously-known-1st-us-death/article_6b198239-2968-540a-a562-9eb1132cb3da.html
I would say that coming up with an exact number is going to be pretty tough. I would say it is reasonable to come up with boundary conditions though. We know the numbers reported on deaths is probably pretty accurate, and that the number of cases is probably under reported due to the lack of testing, and hospital resources. Using those two numbers we can come up with an upper boundary for the US corona virus death rate at 40,000/750,000 = 5.3%.
The lower boundary condition is more in question, but if we assume that 80% of cases are unreported (this seems to jive with what experts are saying), then we would have 750,000/0.2 = 3,750,000 cases. The death rate is much better understood, so we will still use that number. We would have a US coronavirus death rate at 1.07%.
Assuming 90% of cases are unreported results in a death rate of 0.5%.
I think 90% unreported is probably as low as it could be, and lower than most experts think.
I think it is pretty reasonable to believe the death rate is between 0.5% and 5.3%. I don’t consider 0.5% death rate of a disease that is highly infectious to be anything to take lightly.
Happy for you then, better you dealing with the chaos than I!
In that case we agree. Unfortunately, these are the only numbers we have to work with until we find the denominator. And that is likely going to take a long time relative to the speed of this pandemic. I am still going to take numbers we have vs numbers we can’t get. “Bird in the hand…”, “good plan executed today…” and all that. I agree they are significantly less accurate than desired.
I think you’d be surprised. It’s much less likely that the US government can inflate the number once it is published. We have too many researchers not beholden to the federal government. And you will be able to compare across countries, since it will be advantageous to calculate the stats for each region. Even with all the caveats that come with
So I suppose my message would be, don’t get too discouraged or cynical. I think it is very likely we eventually come up with a reliable number, it just won’t be soon. Or range, since most of these indices are usually a spread. I can definitely appreciate the cost-benefit calculations going on regarding the economy etc.
And, as a small government person, I can also appreciate the skepticism and suspicion of power grab by government.
You’ve described the general approach to the CFR question almost exactly. They’ll work different bounding conditions informed by what we do currently know, and run on several different models
I posted it to address people throwing out 0.1%-0.3% death rates (I have no idea what type of math those numbers came from) as that seems to be about 2-5X lower than even the case of 90% of people going unreported (which seems like a reasonable boundary condition, although I doubt it is that extreme).
I think using boundary conditions is about as good as we are going to get for a while.
1 Like
Oh I agree, apologies for being unclear
1 Like
I think the so-called deep state is using the crisis for their agenda but the virus itself is not a hoax. And what shows is impatience brings it back.
I am not an epidemiologist or a virologist and I’m perhaps ignorantly thinking there’s no herd immunity.
Muh freedom! Come and take it!
1 Like
I think Florida’s a pretty good example of why people are on both sides of this. They have about 3% of total cases and 2% of total deaths.
1 Like