DAISIE_loglik_integrate: Integrates the loglikelihood of a single clade across a parameter weighted by a given distribution

Description

Integrates the loglikelihood of a single clade across a parameter weighted by a given distribution

Usage

DAISIE_loglik_integrate(
  pars1,
  pars2,
  brts,
  stac,
  missnumspec,
  CS_version,
  methode,
  abstolint,
  reltolint,
  verbose
)

Value

A loglikelihood value

Arguments

pars1

Vector of model parameters:

pars1[1] corresponds to lambda^c (cladogenesis rate)
pars1[2] corresponds to mu (extinction rate)
pars1[3] corresponds to K (clade-level carrying capacity)
pars1[4] corresponds to gamma (immigration rate)
pars1[5] corresponds to lambda^a (anagenesis rate).

pars2

Vector of settings:

pars2[1] corresponds to res, the maximum number of endemics or non-endemics for which the ODE system is solved; this must be much larger than the actual number for which the probability needs to be calculated.)
pars2[2] corresponds to M, size of the mainland pool, i.e the number of species that can potentially colonize the island.

brts

Numeric vector of branching times

stac

Numeric of Endemicity status

missnumspec

Numeric of missing species

CS_version

a numeric or list. Default is CS_version = list(model = 1, function_to_optimize = 'DAISIE'), but for a relaxed-rate model the list can contain more elements:

model: the CS model to run, options are 1 for single rate DAISIE model, 2 for multi-rate DAISIE, or 0 for IW test model
function_to_optimize: the DAISIE loglikelihood function that will be optimized. Options are: "DAISIE", default, the full DAISIE loglikelihood "DAISIE_approx", an approximate loglikelihood "DAISIE_DE", an exact loglikelkhood for K = Inf based on the D-E approach
integration_method: the method used to do integraion in the relaxed rate model. Options are: 'standard' the default numerical integration 'MC' Monte Carlo integration 'stratified' using quantiles of the gamma distribution
relaxed_par: the parameter to relax (integrate over) in the relaxed rate model. Options are "cladogenesis", "extinction", "carrying_capacity", "immigration", or "anagenesis"
par_sd: standard deviation of the parameter to relax
par_upper_bound upper bound of the parameter to relax
seed: seed of the random number generator in case of 'MC'
sample_size: size of sample in case of 'MC' or 'stratified'
parallel: use parallel computing or not in case of 'MC' or 'stratified'
n_cores: number of cores to use when run in parallel

methode

Method of the ODE-solver. Supported Boost ODEINT solvers (steppers) are: "odeint::runge_kutta_cash_karp54" "odeint::runge_kutta_fehlberg78" "odeint::runge_kutta_dopri5" "odeint::bulirsch_stoer" without odeint::-prefix, ode method is assumed. The default method overall is "lsodes" for DAISIE_ML_CS() and "ode45" from ode() for DAISIE_ML_IW().

abstolint

Numeric absolute tolerance of the integration

reltolint

Numeric relative tolerance of the integration

verbose

A numeric vector of length 1, which in simulations and `DAISIEdataprep()` can be `1` or `0`, where `1` gives intermediate output should be printed. For ML functions a numeric determining if intermediate output should be printed. The default: `0` does not print, `1` prints the initial likelihood and the settings that were selected (which parameters are to be optimised, fixed or shifted), `2` prints the same as `1 and also the intermediate output of the parameters and loglikelihood, while `3` the same as `2` and prints intermediate progress during likelihood computation.