By continuing to use the site, you agree to our use of cookies and to abide by our Terms and Conditions. We in turn value your personal details in accordance with our Privacy Policy.
Please log in or register. Registered visitors get fewer ads.
I'm guessing the TC is the mean value for new cases each day in that 7 day period. So 6 represents 42 +/- ~3.5 new cases for the week.
The RR for 'this' week is simply derived from RR(last) * TC(this) / TC(last), which seems a bit basic to say the least.
My question for one of you smart cookies out there is: Will this model stay proportionate to the number of active cases or will it drift off? i.e. with such a simple model, would an RR of say 100 this week represent exactly the same number of active cases if the RR returns to 100 after a month or two on the other side of a 'wave' for example.
# WE ARE STEALING THE FUTURE FROM OUR CHILDREN --- WE MUST CHANGE COURSE #
Looks like it's a simple 7 day rolling average. If this is the case then it will remain consistent if similar numbers are posted in a few months time. I can't see the TC number you are referring to on that link so I am not sure about that.
I guess using a rolling average makes some sense for smoothing but perhaps a longer period EMA would show spikes quicker given it would weight more recent numbers more highly than earlier numbers. With a RA it takes a while for spikes to show up.
SB
Avatar - IC410 - Tadpoles Nebula
0
HMG Covid map stats explanation? on 10:46 - Dec 17 with 430 views