ISOpureS1.model_optimize.vv.vv_compute_loglikelihood function

Compute loglikelihood relevant to vv for step 1

Computes the part of the loglikelihood function relevant to optimizing vv for step 1.

ISOpureS1.model_optimize.vv.vv_compute_loglikelihood(vv, sum_log_theta, DD)

Arguments

• vv: Kx1 matrix representing the weights of the normal profiles B_i used to make the weighted combination that forms the mean parameter vector for the Dirichlet distribution over m
• sum_log_theta: the column sums of log(theta), a 1xK matrix
• DD: the number of patients (a scalar)

Returns

The negative of the loglikelihood relevant to optimizing vv

Author(s)

Gerald Quon, Catalina Anghel, Francis Nguyen