yuleWindow: Fit Yule Model (Pure Birth) to Temporal Window of Branching...
In laser: Likelihood Analysis of Speciation/Extinction Rates from Phylogenies

Description Usage Arguments Details Value Note Author(s) References See Also Examples

This function fits the Yule model to any temporal window that includes at least one branching time.

1	yuleWindow(x, st1 = x[1], st2 = 0)

`x`	a numeric vector of branching times
`st1`	the start of the interval you wish to examine
`st2`	the end of the interval you wish to examine

'st1' and 'st2' are given in divergence units before present. If we had a set of branching times with an initial divergence 100 mya, yuleWindow(x, 75, 25) would fit the pure birth model to the portion of the tree between 75 and 25 mya. Calling yuleWindow(x, x[1], 0) will fit the model to the entire tree (identical to pureBirth(x)).

Note that st1 must be greater than st2, because they are given in units of divergence before present.

a list with the following components:

`LH`	the log-likelihood at the maximum
`smax`	the speciation rate giving the maximum log-likelihood

This can be used in conjunction with other models in this package to test a priori hypotheses of rate variation. If, for example, it was hypothesized that a particular climatic event occurring T mya shifted diversification rates, you can use yuleWindow(x, x[1], T) and yuleWindow(x, T, 0) to obtain the log-likelihoods for these two temporal windows. The log-likelihoods can then be summed to obtain the likelihood of the full set of branching times under a 2-rate Yule model with an a priori hypothesized rate-shift. This model would only have two free parameters, in contrast to the yule2rate model, with three parameters (and thus, the AIC would be computed as (- 2 * (sum of log-likelihoods) + 4)).

Dan Rabosky drabosky@umich.edu

Nee, S., R. M. May, and P. H. Harvey. 1994b. The reconstructed evolutionary process. Philos. Trans. R. Soc. Lond. B 344:305-311.

Nee, S. 2001. Inferring speciation rates from phylogenies. Evolution 55:661-668.

fitdAICrc, pureBirth, bd

  data(agamids)
  agbtimes <- getBtimes(string = agamids)
  yuleWindow(agbtimes, 0.22, 0.10)
  # fits Yule model to temporal window between 0.22 and 0.10 divergence
  # units before present