Advertisement
Guest User

Untitled

a guest
Feb 16th, 2020
251
0
Never
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 2.10 KB | None | 0 0
  1. >inb4 is not going to follow your retarded equation et similia.
  2. These predictions are actually merely equations that probably CCP is following too due to test kits availability or deliberately published. Does not represent factual accurate data.
  3. Real predictions, OBVIOUSLY, must be calculated using reliable and verifiable data with S.I.R. or S.I.S. models (I'm working in this).
  4. >Salsa on epidemic models and R-naught:
  5. https://statnet.github.io/nme/d1-s2.pdf
  6. https://mysite.science.uottawa.ca/rsmith43/MAT4996/Epidemic.pdf
  7. https://mysite.science.uottawa.ca/rsmith43/MAT3395/R0.pdf
  8. >Sauce on CCP fake numbers
  9. https://www.barrons.com/articles/chinas-economic-data-have-always-raised-questions-its-coronavirus-numbers-do-too-51581622840
  10.  
  11.  
  12. >inb4 you should calculate the fatality/mortality in this or that way et similia.
  13. The case fatality rate (CFR) is diffentt from mortality rate, CFRs are most often used for diseases with discrete, limited time courses, such as outbreaks of acute infections.
  14. >Sushi on case fatality rate, mortality rate and survival rate:
  15. https://www.britannica.com/science/case-fatality-rate
  16. http://osctr.ouhsc.edu/sites/default/files/sites/default/files/docs/berd/BSE%20Modules/Foundations/Module1/Module%201%20Part%20IV%20Notes.pdf
  17. https://www.tandfonline.com/doi/pdf/10.3109/00016925309136688
  18.  
  19.  
  20. >inb4 your data is wrong/incomplete/different from xyz et similia.
  21. I've tracked by myself all the data from BNO, multiple times during the days. All daily datapoints are amended after Chinese official Hubei and National updates, usually at 00:00 GMT.
  22. >Sausage where the data is retreived:
  23. https://bnonews.com/index.php/2020/02/the-latest-coronavirus-cases/
  24. https://twitter.com/BNODesk
  25.  
  26.  
  27. >inb4 your cases prediction is too wide etc.
  28. Cases jumped like 14,000 in one day from a 2,200 average increment. I'm using a Exponential Smoothing (ETS) algorithm so the past cases jump is inevitably widening the confidentiality bounds.
  29. >Chilli on forecast algo:
  30. https://machinelearningmastery.com/exponential-smoothing-for-time-series-forecasting-in-python/
  31. https://en.wikipedia.org/wiki/Exponential_smoothing
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement