init.alpha.gamma.estimation: Initialization the selection strength alpha using robust...

In pbastide/PhylogeneticEM: Automatic Shift Detection using a Phylogenetic EM

Initialization the selection strength alpha using robust estimation

Description

init.alpha.estimation fits (Y_i-Y_j)^2 ~ gamma^2(1-exp(-alpha*d_ij)) for all couples of tips (i,j) that have the same mean, i.e than are not separated by a shift. Shifts are initialized thanks to a lasso (function init.EM.lasso).

Usage

init.alpha.gamma.estimation(
  phylo,
  Y_data,
  nbr_of_shifts,
  times_shared,
  distances_phylo,
  T_tree,
  subtree.list,
  max_triplet_number,
  alpha_known,
  method.init.alpha.estimation,
  tol_EM,
  h_tree,
  miss,
  masque_data,
  independent,
  ...
)

Arguments

phylo	a phylogenetic tree, class phylo.
Y_data	data at the tips.
nbr_of_shifts	number of shifts used in the EM algorithm
distances_phylo	(matrix) : phylogenetic distance, result of function compute_dist_phy

Details

Function robustbase::nlrob is used for the robust fit.

Value

params_init the list of initial parameters to be used, in the right format.

pbastide/PhylogeneticEM documentation built on Feb. 12, 2024, 1:27 a.m.