rate_gls_boot {evolvability}

Bootstrap of the rate_gls model fit

Description

rate_gls_boot performs a parametric bootstrap of a rate_gls model fit.

Usage

rate_gls_boot(
  object,
  n = 10,
  useLFO = TRUE,
  silent = FALSE,
  maxiter = 100,
  tol = 0.001
)

Arguments

object

The output from rate_gls.

n

The number of bootstrap samples

useLFO

logical: when calculating the mean vector of the traits in the 'recent_evol' analysis, should the focal species be left out when calculating the corresponding species' mean. The correct way is to use TRUE, but in practice it has little effect and FALSE will speed up the model fit (particularly useful when bootstrapping).

silent

logical: whether or not the bootstrap iterations should be printed.

maxiter

The maximum number of iterations for updating the GLS.

tol

tolerance for convergence. If the change in 'a' and 'b' is below this limit between the two last iteration, convergence is reached. The change is measured in proportion to the standard deviation of the response for 'a' and the ratio of the standard deviation of the response to the standard deviation of the predictor for 'b'.

Value

A list where the first slot is a table with the original estimates and SE from the GLS fit in the two first columns followed by the bootstrap estimate of the SE and the 2.5%, 50% and 97.5% quantiles of the bootstrap distribution. The second slot contains the complete distribution.

Author(s)

Geir H. Bolstad

Examples

# See the vignette 'Analyzing rates of evolution' and in the help 
# page of rate_gls.