3 \title{Analysis of Diversification with Survival Models}
5 diversi.time(x, census = NULL, censoring.codes = c(1, 0), Tc = NULL)
8 \item{x}{a numeric vector with the branching times.}
9 \item{census}{a vector of the same length than `x' used as an
10 indicator variable; thus, it must have only two values, one coding
11 for accurately known branching times, and the other for censored
12 branching times. This argument can be of any mode (numeric, character,
13 logical), or can even be a factor.}
14 \item{censoring.codes}{a vector of length two giving the codes used
15 for \code{census}: by default 1 (accurately known times) and 0 (censored
16 times). The mode must be the same than the one of \code{census}.}
17 \item{Tc}{a single numeric value specifying the break-point time to
18 fit Model C. If none is provided, then it is set arbitrarily to the
19 mean of the analysed branching times.}
22 This functions fits survival models to a set of branching times, some
23 of them may be known approximately (censored). Three models are
24 fitted, Model A assuming constant diversification, Model B assuming
25 that diversification follows a Weibull law, and Model C assuming that
26 diversification changes with a breakpoint at time `Tc'. The models are
27 fitted by maximum likelihood.
30 The principle of the method is to consider each branching time as an
31 event: if the branching time is accurately known, then it is a failure
32 event; if it is approximately knwon then it is a censoring event. An
33 analogy is thus made between the failure (or hazard) rate estimated by
34 the survival models and the diversification rate of the lineage. Time
35 is here considered from present to past.
37 Model B assumes a monotonically changing diversification rate. The
38 parameter that controls the change of this rate is called beta. If
39 beta is greater than one, then the diversification rate decreases
40 through time; if it is lesser than one, the the rate increases through
41 time. If beta is equal to one, then Model B reduces to Model A.
44 A NULL value is returned, the results are simply printed.
47 Paradis, E. (1997) Assessing temporal variations in diversification
48 rates from phylogenies: estimation and hypothesis
49 testing. \emph{Proceedings of the Royal Society of London. Series
50 B. Biological Sciences}, \bold{264}, 1141--1147.
52 \author{Emmanuel Paradis \email{Emmanuel.Paradis@mpl.ird.fr}}
54 \code{\link{branching.times}}, \code{\link{diversi.gof}}
55 \code{\link{ltt.plot}}, \code{\link{birthdeath}},
56 \code{\link{bd.ext}}, \code{\link{yule}}, \code{\link{yule.cov}}