Quote:
Originally Posted by russellw
One last set of stats for today based on the obvious question of are we winning the war against this pandemic.
The only real measure we have of that is the new infection rates and we saw that the polynomial trend line for Australia (above) has about the sort of curve you'd expect with a gentle ramp up followed by a downward curve.
If I might digress for a moment. For those who don't know, polynomial trend lines are used to smooth out bumps in data sets that can be caused by inconsistent data or weekday vs weekend trends (for example) and they can be 'ordered' to indicate the maximum number of fluctuations or bends that occur on the line. The ones we have been using up to now are 2nd order polynomial trendlines which generally only have one hill or valley.
I've chosen a number of countries that have had high case rates and plotted their polynomial trendlines below:
image
You ideally want to see something like Italy (red) where the downward curve is a bit steeper than say Iran (yellow) but you definitely don't want to see any that are still on the upward trend like the UK and Turkey.
The third order polynomial doesn't show anything much different except it does highlight the impact the measures Germany took (albeit later than they probably should have) has had with a steeper decline than other countries have experienced.
image |
Do you work for the BOM Russellw?
