Rev Language Reference

dnCBDSP(RealPos rootAge, RealPos rootLambda, RealPos rootMu, Distribution__RealPos lambda, Distribution__RealPos mu, RealPos delta, Probability rho, String condition, Taxon[] taxa)

rootAge :	RealPos (pass by const reference)
	The age of the root.
rootLambda :	RealPos (pass by const reference)
	The speciation rate at the root.
rootMu :	RealPos (pass by const reference)
	The extinction rate at the root.
lambda :	Distribution__RealPos (pass by const reference)
	The prior distribution for the speciation rates.
	Default : NULL
mu :	Distribution__RealPos (pass by const reference)
	The prior distribution for the extinction rates.
	Default : NULL
delta :	RealPos (pass by const reference)
	The rate of jumping between rate categories.
rho :	Probability (pass by const reference)
	The taxon sampling probability.
condition :	String (pass by value)
	The condition of the birth-death process.
	Default : survival
	Options : time|survival
taxa :	Taxon[] (pass by value)
	The taxon names used for initialization.

TimeTree

The initial birth and death rates can be specified with the `rootLambda` and `rootMu` arguments, respectively. The rate at which speciation and extinction rate shifts take place is specified by the `delta` argument, and the new speciation or extinction rates are drawn from prior distributions specified in the `lambda` and `mu` arguments. Similar to other birth-death processes in RevBayes, `dnCBDSP` also takes arguments specifying the stopping `condition` of the simulator (either survival or time) and the extant sampling probability `rho`. `dnCBDSP` is very similar to the model implemented in Bayesian Analysis of Macroevolutionary Mixtures (BAMM; Rabosky 2014), particularly in making a strong and potentially problematic assumption that all rate-shift events have been observed (Moore et al. 2016) -- i.e., that no rate shifts are mapped onto unobserved (extinct) branches. For an alternative birth-death-shift model that relaxes this assumption, see `dnCDBDP` (Höhna et al. 2019), which employs a finite number of rate categories instead of drawing rates directly from a continuous distribution.

# draw basic process parameters
taxa <- [taxon("A"), taxon("B"), taxon("C"), taxon("D"), taxon("E")]
root_age ~ dnUniform(0, 2)
root_lambda ~ dnUniform(0, 1)
root_mu ~ dnUniform(0, 1)
sampling_prob <- 1

# simulate tree
tree ~ dnCBDSP(rootAge    = root_age,
               rootLambda = root_lambda,
               rootMu     = root_mu,
               delta      = 0.2,
               rho        = sampling_prob,
               condition  = "survival",
               taxa       = taxa)

• dnCDBDP
• mvBirthDeathEventContinuous
• mvContinuousEventScale
• mvEventTimeBeta
• mvEventTimeSlide