RPANDA (version 2.3)

fitTipData: Maximum likelihood estimators of a model's parameters

Description

Finds the maximum likelihood estimators of the parameters, returns the likelihood and the inferred parameters.

Usage

fitTipData(object, data, error, params0, GLSstyle, v)

Value

value

A numerical value : the lowest -log( likelihood ) value found during the optimization procedure.

inferredParams

The maximum likelihood estimators of the model's parameters.

convergence

An integer code specifying the convergence of the optim function. Please refer to the optim function help files.

Arguments

object

an object of class 'PhenotypicModel'.

data

vector of tip trait data.

error

vector of intraspecific (i.e., tip-level) standard error of the mean. Specify NULL if no error data are available

params0

vector of parameters used to initialize the optimization algorithm. Default value is NULL, in which case the optimization procedure starts with the vector 'params0' specified within the 'model' object.

GLSstyle

boolean specifying the way the mean trait value at the root is estimated. Default value is FALSE in which case the mean at the root is considered as any other parameter. If TRUE, the mean value at the root is estimated with the GLS method, as explained, e.g. in Hansen 1997.

v

boolean specifying the verbose mode. Default value : FALSE.

Author

M Manceau

Details

Warning : This function uses the standard R optimizer "optim". It may not always converge well. Please double check the convergence by trying distinct parameter sets for the initialisation.

References

Manceau M., Lambert A., Morlon H. (2017) A unifying comparative phylogenetic framework including traits coevolving across interacting lineages Systematic Biology

Examples

#Loading an example tree
newick <- "((((A:1,B:0.5):2,(C:3,D:2.5):1):6,E:10.25):2,(F:6.5,G:8.25):3):1;"
tree <- read.tree(text=newick)

#Creating the models
modelBM <- createModel(tree, 'BM')

#Simulating tip traits under the model :
dataBM <- simulateTipData(modelBM, c(0,0,0,1))

#Fitting the model to the data
fitTipData(modelBM, dataBM, v=TRUE)