runTwoStateGibbs: Gibbs sampler for a phylogenetic model with a binary trait.

In vnminin/indorigin: Testing independent/separate origins hypothesis using Bayes factors

Description

Runs a Gibbs sampler for a two-state CMTC on a set of phylogenetic trees, with data at the tips of the trees.

Usage

runTwoStateGibbs(inputTrees, rootDist, traitData, initLambda01, initLambda10,
  priorAlpha01, priorBeta01, priorAlpha10, priorBeta10, mcmcSize, mcmcBurnin,
  mcmcSubsample)

Arguments

inputTrees	multiPhylo object containing a list of trees.
rootDist	Numeric vector of size two with probabilities of the two states at the root of the tree.
traitData	Vector of trait values, coded as 0 and 1.
initLambda01	Initial value of the 0 -> 1 rate.
initLambda10	Initial value of the 1 -> 0 rate.
priorAlpha01	Shape parameter of the Gamma prior on the 0 -> 1 rate.
priorBeta01	Rate parameter of the Gamma prior on the 0 -> 1 rate.
priorAlpha10	Shape parameter of the Gamma prior on the 1 -> 0 rate.
priorBeta10	Rate parameter of the Gamma prior on the 0 -> 1 rate.
mcmcSize	Total number of MCMC iterations.
mcmcBurnin	Number of initial iterations to ignore.
mcmcSubsample	Integer controlling how many MCMC iterations to save. For example, mcmcSubsample = 10 saves every tenth iteration after ignoreing the first mcmcBurnin iterations.

Value

List with two elements: the input trees and a matrix of MCMC output. The columns of MCMC output matrix are Iteration number, Index of a tree that was sampled at this iteration, Log posterior, 0 -> 1 rate, 1 -> 0 rate, Number of 0 -> 1 jumps, Number of 1 -> 0 jumps, Time spent in state 0, Time spent in state 1

vnminin/indorigin documentation built on May 3, 2019, 6:17 p.m.