Thursday, May 21, 2020

Another Useless Model

The NY Times presents, in my opinion, some Columbia Professor's model asserting some politically correct dictum. They state:

Imposing restrictions even a week earlier could have saved lives, researchers found. The coronavirus pandemic has produced a somber Thursday ritual: the tally of unemployment claims across the United States....If the United States had begun imposing social-distancing measures one week earlier in March, about 36,000 fewer people would have died in the pandemic, according to new estimates from Columbia University disease modelers.And if the country had begun locking down cities and limiting social contact on March 1, two weeks earlier than when most people started staying home, a vast majority of the nation’s deaths — about 83 percent — would have been avoided, the researchers estimated.

Since the understanding of just how the virus is transmitted from person to person is not yet well understood, the phenomenological  basis for any model is missing. This is akin to trying to send a man to the moon without knowing Newtons Laws of gravity. I have been there and done that. You can read my first book written in 1969.

Models must reflect reality, must be verifiable, must have a basis in the underlying phenomenon expressed clearly and must have accurate input data. They must give the same result every time and they must be verifiable based upon predictions and actual results. The solution to Newton's three body problem is complex but it got people to and from the moon.

The virus has a strange life. It can adhere to surfaces, get picked up by the hand, transferred to the face when one scratches it under a mask. It is even transferred in fecal matter. Thus any model must perforce of an attempt to be accurate include these factors.

Otherwise we are just playing with spreadsheets and we have just another academic exercise. Shame.