Simulation :
SimulateThomas, SimulateIP, SimulateTypeA, SimulateTypeB
and SimulateTypeC simulate spatial cluster point pattern of Neyman-Scott models
and their extensions. We describe overview of those models briefly in the NScluster documentation ../doc/NScluster-guide.pdf.
Simulation method of each model is described under the corresponding topic.
Parameter estimation :
We adopt the Simplex estimation to maximize the Palm likelihood function (or minimize the negative Palm likelihood function).
The $maximum Palm likelihood estimators$ are called MPLEs, for short.
The Palm intensity function and the analytical form of the Palm log-likelihood of the Tomas model,
Type B model and Type C model are described under the topic SimplexThomas, SimplexTypeB and SimplexTypeC, respectively.
On the other hand, for SimplexIP and SimplexTypeA,
we need to take the alternative form without explicit representation of the Palm intensity function, which need very long c.p.u. time in the minimization procedure.
We parallelize the minimization procedure with OpenMP.
PalmThomas, PalmIP, PalmTypeA, PalmTypeB and PalmTypeC
calculate the non-parametric Palm intensity function estimated directory from a set of point pattern data.
U.Tanaka, Y. Ogata and D. Stoyan, Parameter estimation and model selection for Neyman-Scott point processes, Biometrical Journal, 50, 2008, 43-57.