First, and most importantly, only two papers in the review also check whether other distributions might fit the same data.
I agree this is crucial. Ultimately, lognormal, power law, power law with exponential cutoff, exponential, and stretched exponential distributions cannot be correct. They do not have an upper bound, whereas:
the range of the battle death distribution can’t stretch to infinity
I think it would be worth fitting the data about the number of deaths as a fraction of the total population to truncated versions of the above distributions to the interval between 0 and 1.
Thanks for the post!
I agree this is crucial. Ultimately, lognormal, power law, power law with exponential cutoff, exponential, and stretched exponential distributions cannot be correct. They do not have an upper bound, whereas:
I think it would be worth fitting the data about the number of deaths as a fraction of the total population to truncated versions of the above distributions to the interval between 0 and 1.