I keep seeing it so worth posting this again: why you shouldn't fit disease models to cumulative data http://www.thelancet.com/journals/lancet/article/PIIS0140-6736%2803%2913335-1/fulltext …
-
@c_althaus Noticed in a few places. Don't think it should skew the median R0 estimate too much, but would be issue for forecasting.
@adamjkucharski And it heavily affects confidence intervals. That's why I'm skeptical about the Science paper.