Posts

Showing posts from October, 2022

Covid Update - 10-29-2020

Image
Today's update features the latest model forecast arriving later today. There have been some adjust's in the parameter space the last 24 hours as Reproduction Numbers and the waning of immunity from infection are adjust. This is in response to the latest variant proportions were "officially" published:  Thanks to Dr. Eric Feigl-Ding (EFD) we were able to see this yesterday. Why the delay? These should be posted immediately (after QA/QC) and in my opinion more frequently. The numbers should really be of concern, the change from last week's numbers are dramatic and indicate a faster surge than I expected. Thus my Reproduction number adjustment in the model. I also went back a looked at numbers that provided the impetus to further adjust the post-infection immunity "half-life". When someone tells you that your protection will last 4-6 months ask them if they'd like to buy the Brooklyn Bridge. Semantically it might correct, however a 50% threshold is qui

The Covid Update - 10-28-2022

Image
 Hello, I will be transitioning from Twitter to this page for updates on Covid and other relevant content. The updates will keep rolling throughout the day and be summarized at days end. So let's dig in and start with Dr. Eric Feigl-Ding's (EFD) breaking news from late last evening. I have spent the last couple weeks wondering why MAPS, my model, has been pushing BQ.1 and BQ.1.1 variant infection proportions faster than what I was seeing in the data. This EFD news partially explains why the objective model was just doing it's job. This current wave will be like no previous one. I started formulating things would be much different by observing the BA.5 numbers. You talk about persistent, that variant would just not fade from dominance. This lead me to hypothesize that future sub lineages would exhibit a similar behavior. It quickly ran through my mind that if this trait was combined with an even stronger growth  advantage we would be facing some dire consequences.Here is a s