| Submit your comments on this article | |
| -Signs, Portents, and the Weather- | |
| More magic numbers malarkey... | |
| 2020-05-14 | |
[ZeroHedge] An analysis of computer models used by Professor Neil Ferguson to predict that millions would die of COVID-19- models on which Western leaders had based the ongoing lockdowns - have been torn to shreds. Some have even gone so far as to suggest that his dire models may have something to do with his married lover being an environmental activist.
Mark E Jeftovic, in his Axis of Easy website, says: ’A code review has been undertaken by an anonymous ex-Google software engineer here, who tells us the GitHub repository code has been heavily massaged by Microsoft engineers, and others, in an effort to whip the code into shape to safely expose it to the public. Alas, they seem to have failed and numerous flaws and bugs from the original software persist in the released version. Requests for the unedited version of the original code behind the model have gone unanswered.’" According to Jeftovic, the code produces 'non-deterministic outputs,' which means it will spit out different results for identical inputs - rendering it inappropriate for scientific use. Related: Neil Ferguson: 2020-05-10 Model: Universal Mask-Wearing Would Do More To Reduce Infection Than Indefinite Lockdowns Would Neil Ferguson: 2020-05-10 Destruction of the Imperial College's Ferguson model by some top Google people posting anonymously. Neil Ferguson: 2020-05-07 The War between Experience and Credentials | |
| Posted by:M. Murcek |
| #7 We have to destroy the economy to see what's in the model.... or something like that. |
| Posted by: ruprecht 2020-05-14 14:53 |
| #6 If it saves one life.... |
| Posted by: Procopius2k 2020-05-14 11:26 |
| #5 Hey honey wanna see my |
| Posted by: Lex 2020-05-14 07:32 |
| #4 To paraphrase Dr. Strangelove "When you merely wish to get a predetermined result, there's no limit to the things you can bury in your 'model'" |
| Posted by: M. Murcek 2020-05-14 07:21 |
| #3 models when done correctly, see ^, can help to define what questions to ask and what issues to investigate. GIGO is of particular concern when trying to get a model to provide and answer. |
| Posted by: AlanC 2020-05-14 06:58 |
| #2 The original number crunchers and predictors were astrologers. Just saying. Models can be useful. However, they need clean data, GIGO. They also need clear labeling of confidence. The greater the assumptions and lack of hard data, the lower the predictability and validity. |
| Posted by: Procopius2k 2020-05-14 06:24 |
| #1 Always said numerical simulations are NOT a valid tool of scientific inquiry. |
| Posted by: g(r)omgoru 2020-05-14 02:58 |