Tuesday, October 14, 2014

Hard core statistical predictions

Figure 1:  Data input sheet from the model spreadsheet


If you are serious about looking at the possible direction and magnitude of an Ebola outbreak, you could do a lot worse than starting here.

http://stacks.cdc.gov/view/cdc/24900

CDC has always been the good guys in my mind. Yes, they manage some pretty impressive bureaucratic foul ups off and on, but for the most part they are as good as you get.  The A-team.

Sit down and play with this thing for an hour or two or even a week or so.   Get a feel for how things move around.  See what happens if another R0 case gets through.

Numbers are hard, but working at understanding them and asking questions, lots of questions, is the only way for you to start figuring out which way to jump.


2 comments:

russell1200 said...

This seems to be the scenario for people coming into an unaffected country, and with no additional inputs of people after the first batch. The numbers are still significant (5 digit results) but it is not a worse case scenario.

Degringolade said...

You know, one of the problems with us doomsayers is that we always look at the worst case first.

The reality is, the worst case doesn't always come to pass. In odd occasions, things actually come out all right.

I do hope for such. I just make sure that I am aware of the full spectrum of possibilities and their approximate probabilities.