HZINB_independence: HGZIPS - HZINB (assuming independence)

In sidiwang/hgzips: hgzips

Description

This HZINB_independence function finds hyperparameter estimates by implementing the Expectation-Maximization (EM) algorithm and hierarchical zero-inflated negative binomial model with one gamma component.

Usage

HZINB_independence(
  grid_a,
  grid_b,
  grid_omega,
  init_pi_k,
  init_pi_l,
  init_pi_h,
  dataset,
  iteration,
  Loglik = FALSE,
  zeroes = FALSE,
  N_star = 1
)

Arguments

grid_a	alpha value grid
grid_b	beta value grid
grid_omega	omega value grid
init_pi_k	initial probability of each alpha value for implementing the EM algorithm
init_pi_l	inital probability of each beta value for implementing the EM algorithm
init_pi_h	initial probability of each omega value for implementing the EM algprithm
dataset	a list of squashed datasets that include N_ij, E_ij and weights for each drug (j). This dataset list can be generated by the rawProcessing function in this package.
iteration	number of EM algorithm iterations to run
Loglik	whether to return the loglikelihood of each iteration or not (TRUE or FALSE)
zeroes	A logical scalar specifying if zero counts should be included.
N_star	the minimum Nij count size to be used for hyperparameter estimation. If zeroes are included in Nij vector, please set N_star = NULL HZINB_independence

Value

HZINB_independence a list of estimated probability of each alpha, beta, omega combination and their corresponding loglikelihood (optional)

• theta_EM Estimate of hyperparameters for each EM iteration

• llh logliklihood for each EM iteration (optional)

sidiwang/hgzips documentation built on Jan. 19, 2021, 4:09 p.m.