A Bayesian skyline approach, as suggested by Drummond et al. (2005) and
Billenstein & Höhna (2024), requires as an input a vector of values. These
values can be the same for several consecutive entries. For example, if the
total number of intervals is 4, and we have the two values, 0.2 and 0.5, where
the first value is used once and the second used three times, we could create
a Bayesian skyline vector as `[0.2, 0.5, 0.5, 0.5]`. To construct such
a vector, we use the function `fnBSPInterval`.
The function `fnBSPInterval` takes in two arguments: `x`, the values; and `n`,
the number of times each value is replicated. Currently, the function assumes
that all values are positive real numbers, as `fnBSPInterval` is used for
population sizes in coalescent methods.
a <- [0.1, 0.2, 0.3]
b <- [2, 1, 4]
fnBSPInterval( a, b )