Direct solutions of the master equation

[ChrisRackauckas]

Spectral methods:

https://uu.diva-portal.org/smash/get/diva2:117236/FULLTEXT01.pdf
http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.142.2891
http://www.it.uu.se/research/publications/lic/2006-007/paperB.pdf
https://vtechworks.lib.vt.edu/handle/10919/30018
http://epubs.siam.org/doi/abs/10.1137/070689759
http://ieeexplore.ieee.org/document/6425804/

[ChrisRackauckas]

Additionally, if the maximum number of particles is known (or assumed via truncation), one gets a linear ODE:

https://arxiv.org/abs/1410.1934
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3953644/
https://pdfs.semanticscholar.org/ec67/e296e912f2f1d75534c57e24d758449fb19a.pdf (?)
http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7799362

This then harkens back to disccussions with @jagot on SciML/OrdinaryDiffEq.jl#249 SciML/OrdinaryDiffEq.jl#115 and SciML/OrdinaryDiffEq.jl#116

A lazy A could be constructed for the master equation operator to then solve via expmv! algorithms.

[ChrisRackauckas]

Monomolecular reaction constrained case: https://link.springer.com/article/10.1007%2Fs00285-006-0034-x