Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Politics : Sioux Nation -- Ignore unavailable to you. Want to Upgrade?

To: Arthur Radley who wrote (323911)	3/23/2020 10:51:52 PM
From: Sun Tzu	Respond to of 361804

You can't use the recent data for modeling because you don't know how many infections were there already and you are just discovering them, or that the infection is spreading (it is both, but you are also not testing everyone). The best you can do at this stage is to look at the mortality rate and work backwards from there. Multiply the total number of the dead by 25, and you will have an estimate for the total number of infections a week ago. This is assuming you think it takes an average of 7 days. Some researchers believe there is a 2 weeks lag, in which case the number you get tells you how many were infected two weeks ago. Then you work forward from there. You can use the SIR model (simple enough) and fill in the numbers from scholarly research. Or you can use empirical data (what I did). Keep in mind that over 40% of the infected people do not show any symptoms at all. This is significant if you are going to use proper academic modeling, because the mortality rate is naturally different if your data is excluding asymptomic people (and then you may want to bump up the multiplier to something between 30 - 45). And advantage of using empirical data like I did is that your model will reflect all the flaws in the system. The disadvantage is that as things change, you will have to tune the model parameters. On the other hand, you could have the most perfect epidemic model in the world, but since your data is so imperfect, you will get bad predictions and have to adjust your inputs and assumptions as things change. ============================= I am concerned about FLA. With their elderly population, it can be a real disaster. Hopefully, the virus will hate the sun and the warm weather and will not ravage it. But this is just a hope. So far there is no evidence of this. I am monitoring the progress in TX as a proxy for the effect of the climate. Since Texans are damned before they listen to the Feds, we can assume that a good chunk of change in their infection rate (if any) is due to the warming of the weather.