log: Log transformation of parameter estimates and their...

LogR Documentation

Log transformation of parameter estimates and their uncertainties

Description

Methods for log transforming individual parameter estimates and their uncertainty estimates for use in meta-analytic regression, and then back-transforming mean-log parameter estimates back to mean parameter estimates.

Usage

Log(x,debias=TRUE,...)

Exp(est,VAR.est=0,VAR=0,VAR.VAR=0,variable="area",debias=TRUE,level=0.95,units=TRUE,...)

Arguments

x

A list of UD objects, UD summary objects, or speed objects.

debias

Apply \logχ^2 and \logχ bias corrections if TRUE.

...

Further arguments passed.

est

Point estimate of the mean log-parameter.

VAR.est

Uncertainty in the mean log-parameter estimate (square standard error).

VAR

Variance in the log-parameters.

VAR.VAR

Uncertainty in the log-paramter variance estimate (square standard error).

variable

Variable being back-transformed. Can be "area" or "speed".

level

Confidence level for parameter estimates.

units

Convert result to natural units.

Value

Log returns a list with two slots, log and VAR.log, corresponding to the point estimates and variance estimates of the logged variables.

Exp returns a confidence intervals for the back-transformed mean parameter estimate.

Author(s)

C. H. Fleming.

See Also

meta, mean.

Examples

# load package and data
library(ctmm)
data(buffalo)

# fit movement models
FITS <- AKDES <- list()
for(i in 1:length(buffalo))
{
  GUESS <- ctmm.guess(buffalo[[i]],interactive=FALSE)
  # use ctmm.select unless you are certain that the selected model is OUF
  FITS[[i]] <- ctmm.fit(buffalo[[i]],GUESS)
}

# calculate AKDES on a consistent grid
AKDES <- akde(buffalo,FITS)

# extract 95% areas
AREAS <- lapply(AKDES,summary)

# log transform for further meta-analysis
LOG <- Log(AREAS)

LOG

ctmm documentation built on Nov. 4, 2022, 5:06 p.m.