T_group_path: Title: Tree Path
In Goodgolden/LDTM: Longitudinal Dirichlet Tree Multinomial Model

T_group_path

R Documentation

Title: Tree Path

Description

Title: Tree Path

Usage

T_group_path(
  Y,
  X,
  treeinfo,
  alpha = 0.5,
  cutoff = 0.8,
  model = "dirmult",
  err.conv = 0.001,
  iter.max = 30,
  L.init = NULL,
  lambda.max = NULL,
  lambda.min = NULL
)

Arguments

`Y`	`matrix` of count outcomes
`X`	`matrix` of nxp which is the number of subjects by number of covariates
`treeinfo`	A list objects necessary information from the `tree`. The output has the following properties: Nnodes: `integer` the number of nodes in the given "phylo" tree Ntips: `integer` the number of tips/leaves in the given "phylo" tree Tredge: `matrix` the matrix for the edges information; each row represents the two ends of the edges TreeMat: `matrix` the matrix consists of logical values; the binary value of each taxa (by row) belongs to a given leaf or inner node (by col).
`alpha`	`numeric` the desired lasso parameter. In paper they used (0, 0.25, 0,5, and 1) to investigate the covariate selection. Note: In the paper they noted this as Gamma
`cutoff`	`numeric` the desired cutoff value for the tree path loop
`model`	`character` type of model to use for the Log Likelihood. Options are (Dirichlet Multinomial = "dirmult", Multinomial = "mult", or Dirichlet = "dir")
`err.conv`	`numeric` the desired tolerance level for convergence
`iter.max`	`numeric` maximum number of iterations
`L.init`	`numeric` Initial Lipschitz constant, instead of choosing a constant step size L. We can use the backtracking to choose a suitable L at each iteration. Set to NULL. Note: This is noted at C in Tao Wang and Hongyu Zhao (2017)
`lambda.max`	`numeric` The desired maximum lambda value; set to NULL
`lambda.min`	`numeric` The desired minimum lambda value; set to NULL