EvoRAG (version 2.0)

power.test: performs a simulation based analysis of statistical power

Description

performs a simulation based analysis of statistical power

Usage

power.test(TIME, GRAD, parameters, test.model, threshold_deltaAICc,  REP=1, N, write = "FALSE", wd = "")

Arguments

TIME

vector of evolutionary ages (i.e. node ages ) for sister pair dataset

GRAD

vector of gradient values (i.e. any continuous variable) for sister pair dataset

parameters

A vector listing the model parameters under which to simulate. Model parameters must be in the same order as described in sisterContinuous.

test.model

Any one of the following models are currently supported ("BM_linear", "OU_linear_beta", "OU_linear")

threshold_deltaAICc

A single threshold deltaAICc or a list of such values

REP

How many replicated datasets of TIME and GRAD to use. Default = 1. Example: REP=3 generates a dataset with each element in TIME and GRAD repeated 3 times. This option will be used primarily for calculating statistical power as a function of increasing number of sister pairs

N

The number of simulations to perform

write

If true, writes output to several files

wd

directory to write files to if other than the current working directory. (Windows example, "D:/SIMS/"

Value

Returns a list with the following elements: test.model The model for which power was calculated parameters The parameters under which power was calculated N_sisters The number of sister pairs in the dataset N_sims The number of simulations performed power_test_hypothesis Statistical power calculated for the alternative hypothesis that rates of evolution vary as a linear function of a gradient. Power is returned for each threshold value in threshold_deltaAIC. Where appropriate, power to reject BM_null and OU_null is returned for three comparisons: 1) BMlinear_and_OUlinear_beta_vs_2null: power when simulating data either under BM_linear or OU_linear_beta, but when the OU_linear model is not included in the analysis; 2) BMlinear_and_OUlinear_vs_2null: power when OU_linear_beta is not included; 3) 3gradient_vs_2null: power when all three gradient models are included.power_test_hypothesis The probability of the test model correctly rejecting each of the other null and gradient models on an individual basis.

Details

Performs an analysis of statistical power (e.g. the probability of supporting a true alternative hypothesis) for a given dataset under a given model and set of model parameters. The threshold_deltaAICc should be set at a level that will maintain a type I error (probability of rejecting a true null model) of 0.05. Appropriate threshold_deltaAICc values can be determined using the function TypeI.error. The null hypothesis here tested is that rates of evolution do not vary as a function of gradient (e.g. "BM_null", and "OU_null"). The alternative, is rates do vary as a linear function of a gradient (e.g. "BM_linear", "OU_linear_beta", "OU_linear"). Several hundred or more replicates should be performed. Currently, only "BM_linear", "OU_linear_beta", "OU_linear" are included in the candidate set of gradient models.

See Also

TypeI.error

Examples

## Not run: 
# 
#  ###simulate data
#   set.seed(seed = 3)
#   TIME = runif(n=300, min = 0, max = 10)
#   GRAD = runif(n=300, min = 0, max = 60)
#   DATA1 <- sim.sisters(TIME = TIME, GRAD=GRAD, parameters = c(2, -0.03), 
#      model=c("BM_linear"))
# 
#   ###run power.test
#   model = c("BM_linear")
#   power.test(TIME=TIME, GRAD=GRAD, parameters = c(2, -0.03), test.model="BM_linear", 
#      threshold_deltaAICc = c((1:20)*0.5), REP=1, N=2, write = "FALSE", wd = "")
#  ## End(Not run)#end dontrun