gradients_bootstats: Use bootstrapping to estimate standard errors and confidence...

In MorphoFun/psa: Phenotypic Selection Analysis tools

Description

gradients_bootstats allows the user to calculate the standard deviations and confidence intervals for phenotypic selection gradients that are estimated using the gradients function in psa.

Usage

gradients_bootstats(w, z, conf = 0.95, R = 2000, method = c(1,2))

Arguments

w	Relative fitness.
z	Phenotypic trait(s). Character values are not accepted.
conf	Confidence interval, with 95 percent confidence interval set as a default. See boot.ci for more details.
R	Number of bootstrap replicates. 2000 is set as the default. See ?boot for more details.
method	Method to estimate the selection differential. 1 = matrix algebra approach of phenotypic distributions before and after selection; 2 = ordinary least-squares regression of relative fitness against the trait; "all" = use all of the methods to produce multiple estimates.

Value

gradients_bootstats returns a list of three objects (boot output, standard errors, and confidence intervals).

Output

bootoutput contains the estimates for the phenotypic selection gradients, bias, and standard errors using an "ordinary" resampling method (see the "sim" option in boot::boot for more details)

se contains the bootstrapped standard errors.

ci contains the confidence intervals for four bootstrapping methods (basic, student, percent, and bca). See boot::boot.ci for more details.

Examples

data(BumpusMales)

gradients_bootstats(BumpusMales$w, scale(BumpusMales[,3:11]), method = 1)

MorphoFun/psa documentation built on Nov. 10, 2021, 7:01 a.m.