logllk_ddt_lcm {ddtlcm}

Calculate loglikelihood of the DDT-LCM

Description

Calculate loglikelihood of the DDT-LCM

Usage

logllk_ddt_lcm(
  c,
  Sigma_by_group,
  tree_phylo4d,
  item_membership_list,
  tree_structure_old = NULL,
  dist_mat_old = NULL,
  response_matrix,
  leaf_data,
  prior_class_probability,
  prior_dirichlet,
  ClassItem,
  Class_count
)

Arguments

c

a positive number for the divergence hyperparameter. A larger value implies earlier divergence on the tree

Sigma_by_group

a vector of diffusion variances of G groups

tree_phylo4d

a "phylo4d" object

item_membership_list

a list of G elements, where the g-th element contains the column indices of data corresponding to items in major group g

tree_structure_old

a list of at least named elements: loglikelihoods of the input tree topology and divergence times. These can be directly obtained from the return of this function. Default is NULL. If given a list, then computation of the loglikelihoods will be skipped to save time. This is useful in the Metropolis-Hasting algorithm when the previous proposal is not accepted.

dist_mat_old

a tree-structured covariance matrix from a given tree. Default is NULL.

response_matrix

a N by J binary matrix, where the i,j-th element is the response of item j for individual i

leaf_data

a K by J matrix of logit(theta_kj)

prior_class_probability

a length K vector, where the k-th element is the probability of assigning an individual to class k. It does not have to sum up to 1

prior_dirichlet

a vector of length K. The Dirichlet prior of class probabilities

ClassItem

a K by J matrix, where the k,j-th element counts the number of individuals that belong to class k have a positive response to item j

Class_count

a length K vector, where the k-th element counts the number of individuals belonging to class k

Value

a numeric of loglikelihood

See Also

Other likelihood functions: logllk_ddt(), logllk_div_time_one(), logllk_div_time_two(), logllk_lcm(), logllk_location(), logllk_tree_topology()