confint.moult: Conficence Intervals for Moult Parameters

View source: R/confint.moult.R

confint.moultR Documentation

Conficence Intervals for Moult Parameters

Description

Calculate normal theory and bootstrap percentile intervals for moult parameters. Also returns bootstrap covariance matrix and standard error estimates for parameters.

Usage

## S3 method for class 'moult'
confint(object, parm, level = 0.95, ..., B = 1000, add.plot = TRUE)

Arguments

object

model output returned by call to moult.

parm

currently ignored (should be a vector of numbers or names, specifying for which parameters to calculate confidence intervals).

level

confidence level.

...

additional arguments for plot.

B

integer: number of bootstrap samples to take.

add.plot

logical: should scatterplot of bootstrapped parameters be added?

Details

End date is calculated as mean start date + duration. Half-date is calculated as mean start date + 0.5 * duration.

Value

Returns a list with elements:

bootstrap.distribution

B \times p matrix of bootstrapped parameters.

bootstrap.percentile.ci

Table of bootstrap percentile intervals for all parameters. In addition, confidence intervals for half-date and end date (see details) are given.

bootstrap.vcov

Bootstrap estimates for variances and covariances between all parameters.

bootstrap.SE

Bootstrap estimates of parameter standard errors.

Author(s)

Birgit Erni birgit.erni@uct.ac.za

Examples

data(sanderlings)
m2 <- moult(MIndex ~ Day, data = sanderlings)
## Not run: confint(m2, B = 100) # increase B for better precision

moult documentation built on Aug. 30, 2022, 9:06 a.m.