Lockdown and school closures in Europe may have prevented 3.1m deaths

https://www.imperial.ac.uk/news/198074/ ... yt0mQqZImw

Dr Seth Flaxman, study author from the Department of Mathematics, Imperial College London: ”Using a model based on data from the number of deaths in 11 European countries, it is clear to us that non-pharmaceutical interventions– such as lockdown and school closures, have saved about 3.1 million lives in these countries. Our model suggests that the measures put in place in these countries in March 2020 were successful in controlling the epidemic by driving down the reproduction number and significantly reducing the number of people who would have been infected by the virus SARS-CoV-2.”

Regarding the Covid-19 software the only hope I see is to throw it away and start again. I would have two versions (by two different teams):

1. C++11 and libraries 
2. Python and libraries


All these github versions are a road to nowhere.It is a waste of time and energy.

“Testing does not improve a product; the improving is done by people fixing the bugs that testing has uncovered.”

― Gerald M. Weinberg, Perfect Software: And Other Illusions About Testing

This seems to be the state of play; one group says 80K, the other 200K but root cause seems to elude them.

testing ==> only < 50% complete..

Boris Johnson scrapped a team of senior ministers charged with preparing the country for a pandemic six months before the outbreak of coronavirus in the UK, according to a report.

According to the Daily Mail, the group - called the Threats, Hazards, Resilience and Contingency Committee (THRCC) - was suspended by former prime minister Theresa May on the advice of Cabinet Secretary Sir Mark Sedwill so that civil servants and ministers could focus on Brexit.
The Government committee, which included Cabinet ministers such as Matt Hancock, the Health Secretary, Michael Gove and Gavin Williamson, was abolished by Mr Johnson a few days after he entered Downing Street. The committee was axed without discussing virus control plans.
On Friday night, a former Cabinet minister - who was a member of THRCC - told the Mail that it could have helped to Government to respond more quickly to the virus. 
"Once the pandemic took hold in Italy... alarm bells would have been ringing," the former minister said. "We would have stress-tested the Government’s contingency plans for dealing with a pandemic."

Changing the evolving requirements in mid-stream is not good for models.

BJ was just following Trump, who eliminated the pandemic team and defunded the CDC well ahead of time. Just the conservative playbook. 

BJ was just following Trump, who eliminated the pandemic team and defunded the CDC well ahead of time. Just the conservative playbook. 

I can appreciate that. BJ wants to harmonise health policy with US?, resulting in less frictionless borders.

The Final Countdown

Forecasting for COVID-19 has failed

https://forecasters.org/blog/2020/06/14 ... as-failed/

Been sayin' all along

1.
Many models assume homogeneity, i.e. all people having equal chances of mixing with each other and infecting each other. This is an untenable assumption and in reality, tremendous heterogeneity of exposures and mixing is likely to be the norm. Unless this heterogeneity is recognized, estimated of the proportion of people eventually infected before reaching herd immunity can be markedly inflated
2.Almost all COVID-19 mortality models focused on number of deaths, without considering age structure and comorbidities. This can give very misleading inferences about the burden of disease in terms of quality-adjusted life-years lost, which is far more important than simple death count. For example, the Spanish flu killed young people with average age of 28 and its burden in terms of number of quality-adjusted person-years lost was about 1000-fold higher than the COVID-19 (at least as of June 8, 2020).
3.Complex code can be error-prone and errors can happen even by experienced modelers; using old-fashioned software or languages (e.g. Fortran) can make things worse; lack of sharing code and data (or sharing them late) does not allow detecting and correcting errors
4. Many models are stochastic and need to have a large number of iterations run, perhaps also with appropriate burn-in periods; superficial use may lead to different estimates.
5. For models that use exponentiated variables, small errors may result in major deviations from reality

Lancet editor attacks UK government for ‘catastrophic’ handling of Covid-19 pandemic

https://amp-theguardian-com.cdn.ampproj ... 9-pandemic

I've ordered the book. Will be an "interesting" read..

Lancet editor Leftie activist attacks UK government for ‘catastrophic’ handling of Covid-19 pandemic

https://amp-theguardian-com.cdn.ampproj ... 9-pandemic

I've ordered the book. Will be an "interesting" read..

You're welcome!

Lancet editor attacks UK government for ‘catastrophic’ handling of Covid-19 pandemic

https://amp-theguardian-com.cdn.ampproj ... 9-pandemic

I've ordered the book. Will be an "interesting" read..

In other news about Richard Horton: https://www.theguardian.com/world/2020/jun/04/covid-19-lancet-retracts-paper-that-halted-hydroxychloroquine-trials

Medice, cura te ipsum!

Lancet editor Leftie activist attacks UK government for ‘catastrophic’ handling of Covid-19 pandemic

https://amp-theguardian-com.cdn.ampproj ... 9-pandemic

I've ordered the book. Will be an "interesting" read..

You're welcome!

You're a big Desai fan?

Lancet editor Leftie activist attacks UK government for ‘catastrophic’ handling of Covid-19 pandemic

https://amp-theguardian-com.cdn.ampproj ... 9-pandemic

I've ordered the book. Will be an "interesting" read..

You're welcome!

It's difficult to separate fact  from fiction, science from politics these days. 
I didn't know who RH was before yesterday. Saw him briefly on Newsnight. He came across as a bit surly.

As one of an increasing group of 'armchair epidemiologists' I have an uneasy feeling that the SIR ode models do not cut the mustard. The underlying maths seems to be somewhat anaemic.

correct use stats?

https://www.bbc.com/news/world-53073046

correct use stats?

https://www.bbc.com/news/world-53073046

Nah, bean counting, so it is.