A negative binomial distribution can be used to model various forms of data and allows us to account for clustering, which is an expected clinical phenomena confirmed by our results.
In terms of the concept of independence, the experiment leading to the binomial distribution can be described as follows.
The binomial distribution can be approximated as the Poisson distribution p n, nπ r s 2 / Ω.
When n is sufficiently large (e.g., n≥20), which is generally satisfied by the molecular communication systems, the binomial distribution can be approximated by a normal distribution [15,33].
The negative binomial distribution can be seen as a mixture of Poisson distributions where the median values are drawn from a gamma distribution [17].
In fact, as reported in other studies [ 23, 29, 31, 42, 43], such distributions can often be modeled as negative binomial distributions, which can account for the broad fluctuations of contact durations as well as for the comparatively high fraction of missing links across the chosen classes.
The results proved that the beta-binomial distribution can be very useful for analyzing vegetation landscapes.
In a fixed-effects model, the beta-binomial distribution can be used when there is overdispersion with respect to the binomial distribution.
If additional information on the magnitude of fluctuations were also known, a generalized distribution should be employed, such as a negative binomial [33], which can account for clumping effects (we reproduce all estimations from the main paper using a negative binomial in Supplementary Material S1, Table S2).
