Estimate the nonhomgogenous poisson process intensity function from grouped data

nd_nhpp_fit(
  r,
  n_j,
  d,
  L,
  K,
  J,
  mu_0,
  kappa_0,
  nu_0,
  sigma_0,
  a_alpha,
  b_alpha,
  a_rho,
  b_rho,
  iter_max,
  warm_up,
  thin,
  seed,
  chain,
  num_posterior_samples
)

Arguments

r

vector of distances associatd with different BEFs

n_j

matrix of integers denoting the start and length of each observations associated BEF distances

d

a 1D grid of positive real values over which the differing intensities are evaluated

L

component truncation number

K

intensity cluster truncation number

J

number of rows in r matrix; number of groups

mu_0

normal base measure prior mean

kappa_0

normal base measure prior variance scale

nu_0

inverse chi sqaure base measure prior degrees of freedom

sigma_0

inverse chi square base measure prior scale

a_alpha

hyperparameter for alpha gamma prior

b_alpha

scale hyperparameter for alpha gamma prior

a_rho

hyperparameter for rho gamma prior

b_rho

scale hyperparameter for rho gamma prior

iter_max

total number of iterations for which to run sampler

warm_up

number of iterations for which to burn-in or "warm-up" sampler

thin

number of iterations to thin by

seed

integer with which to initialize random number generator

chain

integer chain label

num_posterior_samples

the total number of posterior samples after burn in

References

Gelman, A., Carlin J., Stern H. and Rubin D. (2004). Bayesian Data Analysis. Cambridge University Press, Chapman & Hall/CRC.

See also

the conjugate normal parameterization in the reference below