[LINK] The government's coronavirus modelling

[Bernard Robertson-Dunn]

On 10/04/2020 8:57 am, Roger Clarke wrote:
> So the first thing that's done is problem-analysis.

Roger I agree with your missive, but would like add to the above comment.

Before you can analyse a problem you have to make sure you are trying to
solve the right problem.

In the case of a government the problem is more than understanding how a
virus might spread. The applied science (i.e. theory) of epidemics is
only a very small part of understanding how a specific virus will spread
in a specific country with a specific health system,  specific rules and
regulations governing behaviours and the impact on the economy and people.

The models that Dr Jansson is comparing don't even include the capacity
of a health system to deal with the case load of SARS-CV2, never mind
all the other factors.

The graphs produced by the models Dr Jansson compares could have been
drawn on the back of an envelop. That's because they are theory. What is
missing from the models is all the messy things that is needed in
real-life models.

Many such models are referenced here:

https://coronavirustechhandbook.com/forecasting

The reality is that governments are not free to make decisions about the
spread of the virus alone. The government's problem is how to achieve
multiple objectives given multiple constrains. To solve this problem
needs multiple, interconnected models fed by real-life, current data. It
also needs the government to realise that it has a control problem, not
a medical/health problem or an economic problem, although both are involved.

To control something you need to understand it. To understand it you
need to describe it. A description is a model, whether it is a written
description, a graph, a mathematical model, or statistics. The skill is
in creating and using the right models 

Which means that when Roger says

> At the end, Jansson write:
>
> > ... I would caution against relying too heavily on modelling ...
>
> Well said, that man!

I'd say, models are critically important, the danger is relying on the
wrong modelling.

-- 

Regards
brd

Bernard Robertson-Dunn
Canberra Australia
email: brd at iimetro.com.au