-
Notifications
You must be signed in to change notification settings - Fork 32
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
out of range doubling/halving central estimate from confidence intervals reported with default but not "vb" method #703
Comments
I think this stems from the difficulty in defining the doubling time as it is a transformed quantity. I believe it changes sign after passing through infinity and not 0 so -23 is in fact contained in these intervals. I think this is probably a doc issue? |
The doubling time is defined in Line 60 in fc92e60
It's calculated from the growth rate So what we're saying is that the doubling time is either a halving time (i.e. it's negative) of more than 4.5 days or a doubling time of more than 7.8 days, with a central estimate being a halving time of 23 days (if growth is 0 then the doubling/halving time is infinite). library("ggplot2")
r <- seq(-0.15, 0.089, by = 0.01)
dt <- log(2) / r
df <- data.frame(r = r, dt = dt)
ggplot(df, aes(x = r, y = dt)) +
geom_line() +
geom_vline(xintercept = -0.03, linetype = "dashed") Created on 2024-07-11 with reprex v2.1.0 |
Summary:
from the epinow() output, doubling/halving estimate, the range of confidence intervals does not include the central estimate.
Description:
I would expect the central estimate to be included in the range. This is visible in the reference documentation of estimate_infection()
I think this could be method-specific, given that the issue appears with the default method, but not with the "vb" method.
Reproducible Steps:
Created on 2024-06-24 with reprex v2.1.0
EpiNow2 Version:
The text was updated successfully, but these errors were encountered: