Sunday, 3 May 2020

You'd be crazy to base policy on the IHME model

The IHME predictions of virus death have been flopping up and down as modellers fiddle with knobs.  Hinderaker at Powerlineblog fills in some blanks.   Good public policy cannot be based on the IHME model.

The numbers change because the model changed, not because the state changed.
A core input is Social distancing but the math isn't explained and effectiveness isn't measured.
The model doesn't try to evaluate partial opening-up of states.

As Hinderaker points out, week by week the predictions of IHME change bizarrely.

Death predictions made week of...  April 11       April 15     April 22
                  South Dakota                 356              181             093
                  North Dakota                 369              032             356
                  Iowa                               743              618             365
                  Minnesota                      442              656             360

It's almost like a random walk. What madness has seized us that elevates modellers and governors  to the place they are today, telling us how to do our jobs and how to run our home life?  Somehow the "flatten the curve so hospital ICU's won't be overwhelmed" has morphed into:  "Disobey me and you'll be sorry."

No comments:

Post a Comment