All the epidemiological models have the same basis

The SIR model is the most basic. It separates the population into 3 groups. (S)usceptible, (I)nfected and (R)emoved. Each has a range 0 to 1. Prior to the outbreak, everyone was in the S group (S=1, I=0, R=0). We currently have a number of people infected and a number in the removed group. Removed is both recovered and deceased. Perhaps currently (S=0.97, I=0.02, R=0.01). This assumes those recovered gain immunity. If they don't then they go back into the S group.

The objective (assuming immunity which was the original accepted viewpoint) is to get S equal to 0 minimising the proportion of R that die.

Now, the models solve two ordinary differential equations to calculate how people move between the groups based on a number of parameters. The transfer from S to I is governed by the R0 value and I. The transfer of people from I to Recovered and deceased are governed by other parameters.

At the start of the pandemic, these parameters were not well defined. They are arguably still not well defined. The modellers made their best guess at the time. As such, different models had differing parameter estimates and therefore very different results.

Now one of the initial models included a behavioural science element. This basically assumed that the population would not tolerate a long term lockdown. The simplest way to get S to 0 minimising the number of deaths is full lockdown until a vaccine would arrive. But if you get a model some dodgy parameters, tell it that we can only lockdown for a limited time and herd immunity is an option, the optimum solution it comes up with is in line with the original model the gov based their approach on.

The Ferguson model doesn't have the behavioural science aspect and had some more reasonable parameter estimates.

I recall mention of 20K deaths as a good outcome. These models are run in a probabilistic manner so give a distribution of outcomes. If 20K was a good outcome, then the median would be substantially more than 20K.

Once this is over and we have a more reliable parameter set we can assess the accuracy of the initial models.

Sorry too nerdy and too many words.

I will make one prediction though: when this is done, the scientists will be thrown under the bus as scapegoats to protect politicians.

Posted By: Timmy_Goat on May 9th 2020 at 13:07:50

Message Thread

Reply to Message

In order to add a post to the WotB Message Board you must be a registered WotB user.

If you are not yet registered then please visit the registration page. You should ensure that their browser is setup to accept cookies.

Log in