That graph looks fishy. Wouldn’t a secondary attack rate of 1 mean that everyone in a cabin with someone sick catches it immediately? Shouldn’t that be deterministically ruled out by the data, and therefore have exactly-zero likelihood?
Also, in general, seeing likelihood graphed on a linear scale makes me think something is very wrong.
attack rate = 1 within a cabin would be everyone catches it at some point (but not necessarily immediately) provided that someone brings it in in the first place—its a rate per sick person rather than per unit time. I don’t have data on whether this is the case although I doubt it.
Technically I suppose having 18 cases in 4-berth cabins does rule that out. My model isn’t sophisticated enough to catch something like that—I look at average illness rate as an input to the binomial distribution, I never check whether the total number is likely. Adding that complexity might help narrow down the true secondary attack rate.
That graph looks fishy. Wouldn’t a secondary attack rate of 1 mean that everyone in a cabin with someone sick catches it immediately? Shouldn’t that be deterministically ruled out by the data, and therefore have exactly-zero likelihood?
Also, in general, seeing likelihood graphed on a linear scale makes me think something is very wrong.
Maybe a bug somewhere?
attack rate = 1 within a cabin would be everyone catches it at some point (but not necessarily immediately) provided that someone brings it in in the first place—its a rate per sick person rather than per unit time. I don’t have data on whether this is the case although I doubt it.
Technically I suppose having 18 cases in 4-berth cabins does rule that out. My model isn’t sophisticated enough to catch something like that—I look at average illness rate as an input to the binomial distribution, I never check whether the total number is likely. Adding that complexity might help narrow down the true secondary attack rate.
I’ve added a log graph.