skygrowth.mcmc.covar: A gibbs-metropolis algorithm for sampling Ne(t) with a 1st...

In mrc-ide/skygrowth: Phylodynamic inference with nonparametric growth rate models

Description

A gibbs-metropolis algorithm for sampling Ne(t) with a 1st order moving average model and using covariate data

Usage

skygrowth.mcmc.covar(
  tre,
  formula,
  data,
  maxSampleTime,
  iter = 1e+05,
  iter0 = 10,
  tau0 = 10,
  tau_logprior = "exponential",
  res = 50,
  beta_logpriors = list(),
  prop_beta_sd = NULL,
  quiet = FALSE,
  control = NULL,
  gamma = NA,
  logRmean = NULL,
  logRsd = 1,
  maxHeight = NULL
)

Arguments

tre	A dated phylogeny in ape::phylo format (see documentation for ape)
formula	An R formula with empty left-hand-side; the right-hand-side specifies relationship of covariates with growth rate of Ne
data	A data.frame, must include 'time' column
maxSampleTime	The scalar time that the most recent sample was collected
iter	iter
iter0	iter0
tau0	Initial guess of the precision parameter
tau_logprior	Prior for precision parameter (character string (gamma or exponential) or function)
res	Number of time points (integer)
beta_logpriors	Optional list of functions providing log density for coefficients (must correspond to data)
prop_beta_sd	Standard deviation of beta proposal kernel
quiet	Provide verbose output?
control	List of options passed to optim
gamma	Death rate. If provided will compute R
logRmean	Mean of R in log space. Determines a lognormal prior on R(t). If used, _gamma_ must be provided
logRsd	SD of R in log space. Determines a lognormal prior on R(t). If used, _gamma_ must be provided
maxHeight	If supplied, will only compute Ne(t) estimates this far back in time. Otherwise will compute to the root of the tree.
mhsteps	Number of mcmc steps

Value

A fitted model including effective size through time

mrc-ide/skygrowth documentation built on May 19, 2020, 5:10 p.m.