emphasis: Expectation-Maximization in Phylogenetic Analysis with...

In franciscorichter/emphasisR: Expectation Maximization on PHylogenetic Applications with Simulations and Importance Sampling

Description

Performs the emphasis framework for parameter estimation of diversification models in phylogenetic trees.

Usage

emphasis(brts,
         model="rpd1",
         init_par,
         soc=2,
         em_tol=0.25,
         sample_size_tol=0.005,
         burnin_sample_size=200,
         burnin_iterations = 20,
         pilot_sample_size=c(200,600))

Arguments

brts	Branching times of the phylogenetic tree
model	Species diversification model
init_par	Initial parameters for emphasis routine
soc	Number of initial species
em_tol	Tolerance levels for the loglikelihood regarding the EM convergenge
sample_size_tol	Tolerance levels for the sample size determination
burnin_sample_size	MC sample size for the burnIn phase
burnin_iterations	Number of initial iterations for the burnIn phase, i.e to drop.
pilot_sample_size	vector with different sample sizes for MCEM routines inside the sample determination phase
paralell	Setting if the MC sampling should be done in paarallel or not

Value

pars	Estimated parameters
MCEM	MCEM chain in a matrix form

Author(s)

F. Richter M.

References

[1] Richter, F., Haegeman, B., Etienne, R. S., & Wit, E. C. (2020). Introducing a general class of species diversification models for phylogenetic trees. Statistica Neerlandica.

Examples

#load data
data(branching_times)

clade = "Bucconidae"
brts =  get(paste0("brts_",clade))
pars = as.numeric(DD_est[DD_est$clade==clade,5:7])

mc = emphasis(brts,model="rpd5c",init_par=c(pars,0))

franciscorichter/emphasisR documentation built on Dec. 20, 2021, 8:50 a.m.