logL.alt.GRW {paleoTSalt}R Documentation

Log-likelihoods for evolutionary models (alternate parameterization)

Description

Returns log-likelihood for general random walk (logL.alt.GRW), unbiased random walk (logL.alt.URW), stasis (logL.alt.Stasis) and OU (logL.alt) models.

Usage

logL.alt.GRW(p, x)
logL.alt.URW(p, x)
logL.alt.Stasis(p, x)
logL.alt.OU(p, x)

Arguments

p a vector of parameters
x a paleoTS object

Details

For the general random walk, p = c(anc, mstep, vstep); for an unbiased random walk, p = c(anc, vstep); for the stasis model, p = c(theta, omega), and for the OU model p = c(anc, vstep, theta, alpha). In general, users will not be access these functions directly, but instead use the optimization functions, which use these functions to find the best-supported parameter values.

Value

The log-likelihood of the parameter estimates (p), given the data (x).

Warning

Because these functions parameterize the models differently, their log-likelihoods are not comparable to those from paleoTS package.

Author(s)

Gene Hunt

References

Hunt, G. 2006. Fitting and comparing models of phyletic evolution: random walks and beyond. Paleobiology32:578–601.

See Also

opt.alt.GRW, logL.GRW

Examples

x<- sim.GRW(ns=20, ms=0, vs=1)
L1<- logL.alt.GRW(p=c(0,0,1), x)        # actual parameters
L2<- logL.alt.GRW(p=c(0,10,1), x)       # should be a bad guess
cat(L1, L2, "\n")

[Package paleoTSalt version 0.1-1 Index]