Hacker News Reader

Comment by cmrivers | Hacker News Reader

The 'tool' used to generate these results is available for download as an excel spreadsheet here: http://stacks.cdc.gov/view/cdc/24900.

Today parameter estimates were published in the NEJM [0] that conflict with the model's default settings. The Eichner incubation distribution, and an infectious duration of maybe 11 days are better inputs, in my opinion.

[0] http://www.nejm.org/doi/full/10.1056/NEJMoa1411100?query=fea...

View on HN

I'm too lazy to input your suggestions, but I'm curious: does it make the scenario better or worse than the CDC suggests?

Thanks for the link, interesting to play with and hats off the the CDC for being so open with their modeling tools. (also, what people do in Excel just blows me away)

I played with your suggested settings and while they are indeed alarming, I'm curious why you think they are better suited. They don't seem to fit the historical curve quite as well as the default CDC ones.

Do you have more background there?

My suggestion is just based on the findings reported in the NEJM paper I linked, which has an incubation of 11 days and a long infectious period.

what people do in Excel just blows me away

It's best to think of it as a programming language unto itself :)

>(also, what people do in Excel just blows me away)

Same. I work in Excel pretty much all day and rock at using it as a swiss army knife. But when people use it to build these hectic models then I'm still thoroughly impressed.

Reply on news.ycombinator.com