@skleinbo @Dan Yeah, the convolution thing works and one can express the characteristic function of the sum of iid truncated Poissons as a simple product of incomplete gamma functions (times a ratio involving only an exponential and a factorial) with first argument an integer. Haven’t tried the inverse transformation, which would be the final step.
Related topics
Topic | Replies | Views | Activity | |
---|---|---|---|---|
Cdf of multivariate normal in distributions.jl | 24 | 10328 | April 24, 2023 | |
Multinomial accepts wrong parameters | 3 | 918 | November 10, 2019 | |
Multinomial(::Int, ::Vector{Any}) error in Turing doing LDA | 5 | 909 | February 9, 2022 | |
Vectorizing observations from multivariate distribution in Turing | 1 | 482 | September 30, 2020 | |
[ANN] MvNormalCDF.jl - multivariate normal CDF | 1 | 830 | November 12, 2021 |