Bioweapon_Modelling_Part_5

Distrusting the Data

Bioweapon Modelling Part 4 Mentioned that progress on the Modelling had been put on hold and that data on younger persons was a particular concern.

It's time for a look at that concern and the numbers as shown in Figure 1.

Figure 1

Three lots of data are under consideration for Modelling, and two datasets are compared to outline the probable extent of the concerns.

The first dataset is the five year average mortality 2015-2019 stratified by age. [1] The data is for 52 weeks beginning in January and includes distortions in the data as driven by holidays and public events. The data is for England and Wales, combined and year totals for each age group are calculated.

The next data set is “Daily deaths by date of occurrence by single year of age and sex, England: 1st June 2014 to 31st July 2022”. [2] To create a near replica of the dataset above, a five year average is calculated. These years begin on the first of November with the first datapoint in November 2014. This change avoids the Christmas and New year distortions, plus this allows a better fit to Modelling excess deaths coincident with a vaccination programme that began in December 2020.

The third dataset is a calculation of excess deaths by week commencing in November 2020. It is derived from the above datasets, plus mortality data from November 2020 onwards.

All of these now draw out attention to the data in Figure 1.

A reasonable assumption is that deaths in England and Wales vs deaths in England should be broadly within the ratios of their respective populations. This does hold for those aged 60 and above. A particular problem is then found in those aged 20-24 and 15-19 where deaths in England, and England only, are higher than the sum of deaths in England and Wales. This makes no sense.

At this point there appears to be no advantage in providing guesses around how many standard deviations are surfacing here, or in speculating what the cause of the differences might be.

About all than can be said is to be unsurprised if errors exceeding one percent are found in the Modelling of the putative Bioweapon's damage to the population.

Said otherwise, do not trust these numbers until separately verified.

[1] https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/deaths/adhocs/11485fiveyearaverageweeklydeathsbysexandagegroupenglandandwalesdeathsoccurringbetween2015and2019

[2] https://www.ons.gov.uk/file?uri=/peoplepopulationandcommunity/birthsdeathsandmarriages/deaths/adhocs/15018dailydeathsbydateofoccurrence1june2014to31july2022bysingleyearofageandsexengland/dailydeathsnov22final.xlsx