power_curves: Power Curves varying delta

In LucyMcGowan/sgpvalue: Functions and examples to understand Second Generation P-values

Description

Power Curves varying delta

Usage

power_curves(delta, n, theta = seq(-3, 3, 0.001), prob = "null")

Arguments

delta	Numeric vector. The width of the indifference zone in SD units.
n	Numeric. The sample size.
theta	Numeric vector. The "true" effect size (i.e., the default is seq(-3, 3, 0.001).
prob	Character. The probability of interest (i.e., if you are interested in the probability that pδ = 0 use null). Must be one of the following: null alternative inconclusive

Value

A plot of the power curves

Examples

## Figure S3 from the second generation P-value paper. The relationship between
## the probability that p_delta = 0 varying delta.

power_curves(delta = c(0, 1/30, 1/2, 1), n = 10, prob = "null")

## Figure S5 from the second generation P-value paper. The relationship
## between probability of data supported compatibility with the null hypothesis,
## and various deltas

## n = 40 (Figure S5 left)
power_curves(delta = c(0, 1/30, 1/2, 1), n = 40, prob = "alternative")
## n = 200 (Figure S5 right)
power_curves(delta = c(0, 1/30, 1/2, 1), n = 200, prob = "alternative")

## Figure S6 from the second generation P-value paper. The relationship between
## the probability of an inconclusive result and various deltas.

## n = 20 (Figure S6 left)
power_curves(delta = c(0, 1/30, 1/2, 1), n = 20, prob = "inconclusive")
## n = 200 (Figure S6 right)
power_curves(delta = c(0, 1/30, 1/2, 1), n = 200, prob = "inconclusive")

LucyMcGowan/sgpvalue documentation built on May 3, 2019, 4:30 p.m.