Neil Ferguson Disease Model is Garbage

The COVID-19 disease model from Imperial College shocked politicians around the world into shutting down economies and throwing hundreds of millions of people out of work. It predicted millions of CoV-19 deaths in the US alone. Its author, Neil Ferguson, was recently forced to resign from a government post as a result of breaking his own “shelter in place” recommendations to carry on a liason with a married woman. If you think that is sleazy, consider the following takedown of the “upgraded” computer code for his disease model:

Imperial finally released a derivative of Ferguson’s code. I figured I’d do a review of it and send you some of the things I noticed. I don’t know your background so apologies if some of this is pitched at the wrong level. It isn’t the code Ferguson ran to produce his famous Report 9. What’s been released on GitHub is a heavily modified derivative of it, after having been upgraded for over a month by a team from Microsoft and others. This revised codebase is split into multiple files for legibility and written in C++, whereas the original program was “a single 15,000 line file that had been worked on for a decade” (this is considered extremely poor practice). A request for the original code was made 8 days ago but ignored, and it will probably take some kind of legal compulsion to make them release it. Clearly, Imperial are too embarrassed by the state of it ever to release it of their own free will, which is unacceptable given that it was paid for by the taxpayer and belongs to them.

Due to bugs, the code can produce very different results given identical inputs. They routinely act as if this is unimportant.

This problem makes the code unusable for scientific purposes, given that a key part of the scientific method is the ability to replicate results. Without replication, the findings might not be real at all – as the field of psychology has been finding out to its cost. Even if their original code was released, it’s apparent that the same numbers as in Report 9 might not come out of it.

In issue 116 a UK “red team” at Edinburgh University reports that they tried to use a mode that stores data tables in a more efficient format for faster loading, and discovered – to their surprise – that the resulting predictions varied by around 80,000 deaths after 80 days…

Imperial admit there’s a bug by referencing a code change they’ve made that fixes it. The explanation given is “It looks like historically the second pair of seeds had been used at this point, to make the runs identical regardless of how the network was made, but that this had been changed when seed-resetting was implemented”. In other words, in the process of changing the model they made it non-replicable and never noticed.

Why didn’t they notice? Because their code is so deeply riddled with similar bugs and they struggled so much to fix them that they got into the habit of simply averaging the results of multiple runs to cover it up… and eventually this behaviour became normalised within the team.

In issue #30, someone reports that the model produces different outputs depending on what kind of computer it’s run on (regardless of the number of CPUs). Again, the explanation is that although this new problem “will just add to the issues” … “This isn’t a problem running the model in full as it is stochastic anyway”.

Although the academic on those threads isn’t Neil Ferguson, he is well aware that the code is filled with bugs that create random results. In change #107 he authored he comments: “It includes fixes to InitModel to ensure deterministic runs with holidays enabled”. In change #158 he describes the change only as “A lot of small changes, some critical to determinacy”.

Imperial are trying to have their cake and eat it. Reports of random results are dismissed with responses like “that’s not a problem, just run it a lot of times and take the average”, but at the same time, they’re fixing such bugs when they find them. They know their code can’t withstand scrutiny, so they hid it until professionals had a chance to fix it, but the damage from over a decade of amateur hobby programming is so extensive that even Microsoft were unable to make it run right.

… All papers based on this code should be retracted immediately. Imperial’s modelling efforts should be reset with a new team that isn’t under Professor Ferguson, and which has a commitment to replicable results with published code from day one. __ Ferguson Code Review

And so on, follow the link above and read the whole thing. The world-devastating Imperial College disease model is crap.

Sadly, modelers at the University of Washington seem just as eager to use sensationalist public pronouncements to sway policy as was Neil Ferguson’s team at Imperial. But fortunately, many other epidemiologic modelers are far more modest:

The point of flattening the curve isn’t to eliminate transmissions. It’s to let the transmissions happen slowly. Unless we get lucky, we are going to have a similar amount of cases, the same area under the curve. It’s a matter of pacing these cases out over the next 3 months so that our healthcare system can handle them.”

… In reality, we don’t know when we go back to our normal lives after social distancing, the rate of infection will go up or not. We don’t know whether we will have a spike in the number of cases again. __

The lockdowns that resulted from the “Imperial model panic” are not cost-free. Sadly, politicians and newly empowered public health officials and academics do not seem to understand the life or death tradeoffs involved. Nor are they qualified to deal with them if they were aware.

Crappy Computer Models have Real World Consequences

Apparently the majority of voters in western countries were panicked enough to allow their democratically elected leaders to imprison them in their homes, and take away their ability to provide for themselves and their loved ones. But now a large proportion of them are reconsidering the aftermath of these draconian lockdowns.

73% say that it’s important for their mental wellbeing to be able to see people face-to-face again. 61% are concerned about the health risks associated with prolonged isolation.

The awareness comes from firsthand experience. 33% of voters have close friends or family members who have been severely depressed during the lockdown. Additionally, 23% know people close to them who have been drinking too much. On a personal level, 35% have gained weight or experienced other health related problems as a result of the shutdown.

Many people with heart disease, cancer, and other potentially fatal illnesses have been unwilling or unable to get vital therapies or interventions as a result of the media/political hysteria over Wuhan CoV-19.

There will be excess deaths with lockdowns and there will be excess deaths without lockdowns. It is not as if one approach will lead to the saving of more lives. Over the long run, many more will die as a result of lockdowns than would be saved by them.

Lives are being lost to the lockdown, a toll that will mount the longer the economy remains shuttered. Some doctors estimate that the closure of hospitals to non-coronavirus cases and the reluctance of patients to burden 911 have increased mortality as much as the virus. The global depression will devastate life expectancies in the less-developed world. Overdose deaths and suicides brought on by joblessness and loss of hope will rise, as more and more businesses fold permanently.

The rhetoric of lockdown proponents is growing more apocalyptic. “A Virus Tightens Its Deadly Grip” announced the lead print headline in Wednesday’s New York Times — even as the data keep reinforcing the case against universal shutdown. Infection outbreaks are occurring in highly specific locales, not universally: nursing homes, meatpacking plants and prisons. Deaths are tragically concentrated in the former.

On Wednesday, Cuomo announced the “shocking” news that 84 percent of all hospital admissions were either people sheltering at home or nursing home residents. He shouldn’t have been surprised. The risk of coronavirus infection occurs overwhelmingly indoors. Researchers in China identified only one outdoor outbreak of infection among over a thousand cases studied. Most transmissions occurred at home, rendering the close-down-all-businesses-and-shelter-in-place rule contraindicated. __

Should people wear masks in public?

I recommend wearing a mask for shopping and other close activities in the public sphere. But I do not recommend that authorities throw people in jail for not wearing masks. A lot of people inside government are allowing all of this newfound power to go to their heads. If they are not more careful, those heads may find a different resting place than their shoulders. 😉

Big government figureheads are always looking for an opportunity to expand the size of corrupt government. Call it job security. But there are some times when it would be best to forego the temptation.

Robots Work 24/7 and Don’t Get Sick

The Wuhan CoV-19 government debacle is adding more incentive for employers to speed up the switch-over from human employees to robots:

JBS SA, the world’s largest meat producer, is preparing to install robots in slaughterhouses to mitigate the spread of COVID-19 among human employees working on the production line.

JBS SA CFO Guilherme Cavalcanti recently said the Brazilian processing company expects to expand automation at its facilities across the world.

… CNN reports that grocers – big and small chains alike – are turning to robots for performing various tasks like cleaning floors, stocking shelves and delivering groceries to shoppers. The covid crisis could even prompt online retail warehouses like Amazon to invest more into automation technology as well.

The New York Times reported that the outbreak is boosting demand for Zhen Robotics and its RoboPony, a self-driving cart that is sold to retailers, hospitals, malls and apartment complexes.

A group of scientists on the editorial board of Science Robotics are further calling for robots to do the “dull, dirty, and dangerous jobs” of infectious disease management by replacing certain hospital jobs like disinfecting robots combing rooms/floors and working in labs.

A recent report by A3, Association For Advancing Automation, further details all the ways that artificial intelligence and automation is being used in different industries to combat the coronavirus. __ ZeroHedge and Source

Perhaps half of all small businesses in the US will disappear as a result of this lockdown madness. That will lead to significant hardship. But perhaps the saddest outcome of this political disaster is that most of the citizens who passively went along with the dictatorial stripping of their freedoms, will never understand that they were ever wrong, or what an intelligent and free person might have done differently.

This entry was posted in Biomedicine, Pandemic and tagged . Bookmark the permalink.