mcmc_pwc: RJMCMC: MCMC.

In JPNotts/Package: Simulate Spike Sequences

Description

RJMCMC: MCMC.

Usage

mcmc_pwc(
  d.spikes,
  end.time,
  iter,
  burn,
  k.max,
  lambda,
  kappa,
  mu,
  kappa0,
  hyper.param = c(1, 0.001),
  sigma.h = NULL,
  start.hyper = 1000,
  hyper.initial = 0.1,
  T.min.param = NULL,
  T.min.initial = 0.05,
  sigma.t = NULL,
  ISI.type = "Gamma",
  do.log = TRUE,
  show.iter = FALSE,
  which.heights = "independent"
)

Arguments

d.spikes	Data frame containing spike sequences.
end.time	End time of the experiment
iter	Number of iterations to record
burn	Number of iterations to burn before recording.
k.max	Maximum number of change points in RJMCMC.
lambda	Parameter of the Poisson distribution.
kappa	parameter of priors height
mu	parameter of priors height
kappa0	parameter of priors height
hyper.param	Value of the hyper parameter, or values for the gamma prior of the hyper parameter.
sigma.h	Sigma used in RW-Metropolis for the ISI parameter.
start.hyper	Iteration number to begin the ISI parameter.
hyper.initial	Initial value of the ISI parameter.
T.min.param	Value of the refractory period, or values for the gamma prior of the refractory period parameter.
T.min.initial	Initial value of the refractory period.
sigma.t	Sigma used in RW-Metropolis for the refractory period parameter.
ISI.type	The ISI distribution.
do.log	Flag, where if do.log is true the calculations are computed on the log scale.
show.iter	Flag for whether to print the number of iterations complete into the console.
which.heights	Method for prior distribution of heights, either 'independent' or 'martingale'.

Value

Iterations of the MCMC algorithm.

JPNotts/Package documentation built on Oct. 5, 2021, 2:04 p.m.