get_perEvent_probs: Get the per-event probabilities at cladogenesis

In BioGeoBEARS: BioGeography with Bayesian (and Likelihood) Evolutionary Analysis in R Scripts

Description

At a cladogenesis event, a large number of events are possible. The simplest way to compute these is just to assign some weight to each event, then sum all the events and divide by the sum to get the probabilities. More complex schemes can be imagined, but these are fairly pointless as they would all break down once e.g. distance-dependence, user-specified connectivities, etc., are imposed.

Usage

  get_perEvent_probs(params_table, sumval = 1,
    plotwhat = "est")

Arguments

params_table	The params_table from a BioGeoBEARS_model_object.
sumval	Default=1.
plotwhat	Default "est", use "init" to get the initial starting values instead.

Details

In addition, one could imagine trying to assign total probabilities to each category of event, but each row of the cladogenesis matrix may have a different count of the different types of events (one row may have 1 y event and 2 j events; another row may have 4 j, 2 v, and 2 s, and 0 y events; etc.).

One thing that IS meaningful is the per-event weight, i.e. the values that the program is using for j, v, y, and s. These ARE meaningful, as long as they are forced to sum to some value (default 4). This ensures that they are identifiable (otherwise, j,v,y,s=1 and j,v,y,s=2 would be the same model).

This function calculates the per-event weight as a proportion of some total weight, e.g. default 1. If the optim result was j=0, s=1, y=1, v=1, the get_perEventprobs() result would be 0, 0.333, 0.333, 0.333.

Value

wts Return the per-event weights

Note

Go BEARS!

Author(s)

Nicholas J. Matzke [e-mail]

References

http://phylo.wikidot.com/matzke-2013-international-biogeography-society-poster

Matzke_2012_IBS

See Also

rbind

Examples

# default DEC+J model
BioGeoBEARS_run_object = define_BioGeoBEARS_run()
BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table
params_table = BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table
params_table

get_perEvent_probs(params_table)


# DEC+J model
BioGeoBEARS_run_object = define_BioGeoBEARS_run()
BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["j","type"] = "free"
BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["j","init"] = 1
BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["j","est"] = 1

BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table

BioGeoBEARS_run_object$BioGeoBEARS_model_object =
calc_linked_params_BioGeoBEARS_model_object(
BioGeoBEARS_model_object=BioGeoBEARS_run_object$BioGeoBEARS_model_object,
update_init=TRUE)


BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table
params_table = BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table

get_perEvent_probs(params_table)

BioGeoBEARS documentation built on May 29, 2017, 8:36 p.m.