ratebystate: Method for investigating the rate of one trait as a function...

In phytools: Phylogenetic Tools for Comparative Biology (and Other Things)

Method for investigating the rate of one trait as a function of the state of another

Description

Statistical test of whether the rate of a continuous character might be influenced by the state of another.

Usage

ratebystate(tree, x, y, nsim=100, corr=c("pearson","spearman"), ...)

Arguments

tree	phylogenetic tree.
x	a continuous character - the dependent variable in the model.
y	a second continuous trait - the response variable.
nsim	number of simulations for hypothesis testing.
corr	correlation method to use. Same as in cor.
...	optional arguments which include sim.method ("fastBM" or "sim.corrs"; see fastBM and sim.corrs); method ("by.node" or "by.branch" indicating whether to assume the rate varies as a function of the node state or the mean branch state); message - a logical value indicating whether or not to return corr and method; finally logarithm - indicating whether or not to fit a model in which the variance of Brownian evolution in y changes as a multiplicative function of x. The default is logarithm=FALSE.

Details

This function attempts to ask if the rate of a continuous character, y, depends on the state of a separate continuous trait, x. This is accomplished by regressing the squared contrasts in y on the branch or node ancestral estimates of x.

Value

This function returns an object of class "ratebystate" with up to the following four elements:

beta	value of the regression coefficient for square of the contrasts in y regressed on the ancestral or branch-wise estimated states for x.
r	correlation coefficient for corr=corr.
corr	string giving the value of corr.
method	string giving the value of method.

Author(s)

Liam Revell [e-mail]

References

Revell, L. J. (2024) phytools 2.0: an updated R ecosystem for phylogenetic comparative methods (and other things). PeerJ, 12, e16505.

See Also

fastAnc, pic

phytools documentation built on April 3, 2025, 10:43 p.m.