find.mle_FPK_multiple_clades_same_V_different_sig2: Maximum-likelihood estimation

In BBMV: Models for Continuous Traits Evolving in Macroevolutionary Landscapes of any Shape

Description

Maximum-likelihood estimation of the FPK model on multiple clades at once.

Usage

find.mle_FPK_multiple_clades_same_V_different_sig2(model, 
  method = "Nelder-Mead", init.optim = NULL)
find.mle_FPK_multiple_clades_same_V_same_sig2(model, 
  method = "Nelder-Mead", init.optim = NULL)

Arguments

model	An FPK or BBMV model fitted to multiple clades, as generated by lnl_FPK_multiclades_same_V_different_sig2, lnl_BBMV_multiclades_same_V_different_sig2, lnl_BBMV_multiclades_same_V_same_sig2, or lnl_FPK_multiclades_same_V_same_sig2.
method	The optimization routine to be used: can be either "Nelder-Mead" (the default) or "L-BFGS-B". See the documentation of the optim function for more details. From our experience, "Nelder-Mead" seems to produce better results.
init.optim	A vector of initial values for model parameters to start the optimization algorithm. If left NULL (as is by default), the function chooses a reasonable starting point, but you might want to play around with it.

Author(s)

F. C. Boucher

Examples

## Not run: 
# We first create a potential that we will use to simulate trait evolution 
# It has two peaks of very unequal heights
x=seq(from=-1.5,to=1.5,length.out=100)
bounds=c(min(x),max(x)) # the bounds we use for simulating
a=8 ; b=-4 ; c=1
V6=a*x^4+b*(x^2)+c*x
step_size=(max(bounds)-min(bounds))/(100-1)
V6_norm=exp(-V6)/sum(exp(-V6)*step_size) # the step size on the grid
par(mfrow=c(1,1))
plot(V6_norm,type='l')

# Now we simulate a tree and a continuous trait for 3 independent clades. 
#The trait evolves in the same macroevolutionary landscape but with different evolutionary rates.
tree=sim.bdtree(stop='taxa',n=25)
tree$edge.length=100*tree$edge.length/max(branching.times(tree))
TRAIT= Sim_FPK(tree,x0=0.5,V=V6,sigma=1,bounds=bounds)
tree1=tree ; TRAIT1=TRAIT

tree=sim.bdtree(stop='taxa',n=25) 
tree$edge.length=100*tree$edge.length/max(branching.times(tree))
TRAIT= Sim_FPK(tree,x0=0.5,V=V6,sigma=0.5,bounds=bounds) 
tree2=tree ; TRAIT2=TRAIT

tree=sim.bdtree(stop='taxa',n=25) 
tree$edge.length=100*tree$edge.length/max(branching.times(tree))
TRAIT= Sim_FPK(tree,x0=0.5,V=V6,sigma=0.1,bounds=bounds) 
tree3=tree ; TRAIT3=TRAIT
rm(tree) ; rm(TRAIT)

TREES=list(tree1,tree2,tree3)
TRAITS=list(TRAIT1,TRAIT2,TRAIT3)

# Fit the FPK model using ML
testbFPK4=lnl_FPK_multiclades_same_V_different_sig2(trees=TREES,
  traits=TRAITS,a=NULL,b=NULL,c=NULL,Npts=50)
fitbFPK4=find.mle_FPK_multiple_clades_same_V_different_sig2(model=testbFPK4,
  method='Nelder-Mead',init.optim=NULL)
  
## End(Not run)

BBMV documentation built on May 1, 2019, 10:26 p.m.