norm_pval: Visualization of p-values for basic hypothesis tests with the...

In EmanuelSommer/PvalVis: A quick and easy way of visualizing p-values for basic hypothesis tests

Description

Given z ~ N(μ,σ^2) the function calculates the p-value and visualizes the result as the area under the density function. Furthermore the mean and the values one and two standard deviations from the mean are highlighted.

Usage

norm_pval(z_value = 0, mean = 0, sd = 1, direction = c("extreme",
  "less", "greater", "both"))

Arguments

z_value	The value of a test statistic with the underlying normal distribution (values that are very far away from the mean - roughly more than 4 times the standard deviation - are not recommend to use as the p-value will be approximately 0 or 1 anyways)
mean	The mean of the underlying normal distribution (default is 0).
sd	The standard deviation of the underlying normal distribution (default is 1).
direction	The 'direction' of the test with respect to z: extreme (default) The p-value will be calculated using min(P(X ≤ z),P(X ≥ z)) with X~N(μ,σ^2) less The p-value will be calculated using P(X ≤ z) with X~N(μ,σ^2) greater The p-value will be calculated using P(X ≥ z) with X~N(μ,σ^2) both The p-value will be calculated using 2*min(P(X ≤ z),P(X ≥ z)) with X~N(μ,σ^2) So for the first three options a one sided hypothesis gets tested and for the last one a two sided hypothesis is tested.

Value

a ggplot2 object displaying the results

Author(s)

Emanuel Sommer

Examples

norm_pval(-2)
norm_pval(z_value = 100, mean = 120, sd = 20, direction = "both")

EmanuelSommer/PvalVis documentation built on Nov. 20, 2020, 1:34 a.m.