opt.joint.GRW {paleoTS}

Fit evolutionary models using the "Joint" parameterization

Description

Fit evolutionary models using the "Joint" parameterization

Usage

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

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

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

opt.joint.StrictStasis(y, pool = TRUE, cl = list(fnscale = -1), hess = FALSE)

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

Details

These functions use the joint distribution of population means to fit models using a full maximum-likelihood approach. This approach was found to have somewhat better performance than the "AD" approach, especially for noisy trends (Hunt, 2008).

Value

a paleoTSfit object with the model fitting results

Functions

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

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

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

Note

It is easier to use the convenience function fitSimple.

References

Hunt, G., M. J. Hopkins and S. Lidgard. 2015. Simple versus complex models of trait evolution and stasis as a response to environmental change. Proc. Natl. Acad. Sci. USA 112(16): 4885-4890.

See Also

fitSimple, opt.GRW

Examples

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