paleoTS (version 0.5.3)

opt.GRW: Fit evolutionary model using "AD" parameterization

Description

Fit evolutionary model using "AD" parameterization

Usage

opt.GRW(
  y,
  pool = TRUE,
  cl = list(fnscale = -1),
  meth = "L-BFGS-B",
  hess = FALSE
)
opt.URW(
  y,
  pool = TRUE,
  cl = list(fnscale = -1),
  meth = "L-BFGS-B",
  hess = FALSE
)
opt.Stasis(
  y,
  pool = TRUE,
  cl = list(fnscale = -1),
  meth = "L-BFGS-B",
  hess = FALSE
)
opt.StrictStasis(
  y,
  pool = TRUE,
  cl = list(fnscale = -1),
  meth = "L-BFGS-B",
  hess = FALSE
)

Value

a paleoTSfit object with the model fitting results

Arguments

y

a paleoTS object

pool

if TRUE, sample variances are substituted with their pooled estimate

cl

optional control list, passed to optim()

meth

optimization algorithm, passed to optim()

hess

if TRUE, return standard errors of parameter estimates from the hessian matrix

Functions

• opt.URW(): fit the URW model by the AD parameterization

• opt.Stasis(): fit the Stasis model by the AD parameterization

• opt.StrictStasis(): fit the Strict Stasis model by the AD parameterization

Details

These functions use differences between consecutive populations in the time series in order to remove temporal autocorrelation. This is referred to as the "Ancestor-Descendant" or "AD" parameterization by Hunt [2008], and it is a REML approach (like phylogenetic independent contrasts). A full ML approach, called "Joint" was found to have somewhat better performance (Hunt, 2008) and generally should be used instead.

References

Hunt, G. 2006. Fitting and comparing models of phyletic evolution: random walks and beyond. Paleobiology 32(4): 578-601.

See Also

fitSimple, opt.joint.GRW

Examples

x <- sim.GRW(ns = 20, ms = 1)  # strong trend
plot(x)
w.grw <- opt.GRW(x)
w.urw <- opt.URW(x)
compareModels(w.grw, w.urw)