MCMC_component {seeds}

Componentwise Adapted Metropolis Hastings Sampler

Description

Algorithm implemented according to Engelhardt et al. 2017.

Usage

MCMC_component(
  LOGLIKELIHOOD_func,
  STEP_SIZE,
  STEP_SIZE_INNER,
  EPSILON,
  JUMP_SCALE,
  STEP,
  OBSERVATIONS,
  Y0,
  INPUTDATA,
  PARAMETER,
  EPSILON_ACT,
  SIGMA,
  DIAG,
  GIBBS_par,
  N,
  BURNIN,
  objective
)

Arguments

LOGLIKELIHOOD_func

likelihood function

STEP_SIZE

number of samples per mcmc step. This should be greater than numberStates*500.Values have direct influence on the runtime.

STEP_SIZE_INNER

number of inner samples. This should be greater 15 to guarantee a reasonable exploration of the sample space. Values have direct influnce on the runtime.

EPSILON

vector of hidden influences (placeholder for customized version)

JUMP_SCALE

ODE system

STEP

time step of the sample algorithm corresponding to the given vector of time points

OBSERVATIONS

observed state dynamics e.g. protein concentrations

Y0

initial values of the system

INPUTDATA

discrete input function e.g. stimuli

PARAMETER

model parameters estimates

EPSILON_ACT

vector of current hidden influences

SIGMA

current variance of the prior for the hidden influences (calculated during the Gibbs update)

DIAG

diagonal weight matrix of the current Gibbs step

GIBBS_par

GIBBS_PAR[["BETA"]] and GIBBS_PAR[["ALPHA"]]; prespecified or calculated vector of state weights

N

number of system states

BURNIN

number of dismissed samples during burn-in

objective

objective function

Details

The function can be replaced by an user defined version if necessary

Value

A matrix with the sampled hidden inputs (row-wise)