calc_power: Calculate power for the noninferiority null

In li-xinran/RIQITE: Randomization Inference for Quantiles of Individual Treatment Effects

Calculate power for the noninferiority null

Description

Given a specified DGP for control outcomes and treatment impacts, calculate power to detect effects. Calculates superpopulation power via simulation, where we simulate series of datasets of size n, randomized into treatment and control, and then test for detection of effects. Function returns proportion of times effects are detected.

Usage

calc_power(
  n,
  Y0_distribution = rnorm,
  tx_function = "constant",
  p_tx = 0.5,
  R = 100,
  iter_per_set = 10,
  percentile = 1,
  alpha = 0.05,
  c = 0,
  k.vec = NULL,
  alternative = "greater",
  method.list = list(name = "Stephenson", s = 10),
  targeted_power = TRUE,
  use_pval_bound = FALSE,
  score = NULL,
  stat.null = NULL,
  nperm = 1000,
  calc_ICC = FALSE,
  quantile_n_CI = NA,
  summarise = TRUE,
  ...
)

Arguments

n	Sample size to test
Y0_distribution	Either a function to generate a set of Y0 values, or a list of discrete values to bootstrap sample from (with replacement). Function takes single parameter of sample size. Default is normal.
tx_function	Either a string name of an implemented distribution (rexp, rnorm, linear, etc.) or a function that takes Y0 and returns a set of treatment effects. Default is a constant tx effect. tx_constant.
R	Number of permutation replicates for power calculation.
iter_per_set	Number of permutations per finite dataset.
percentile	Which quantile is of interest (default is 1, the max).
alpha	Alpha for testing (default of 0.05)
c	Threshold for null (see pval_quantile)
alternative	A character takes value "greater", "less" and "two.sided", indicating the alternative hypothesis.
method.list	A list specifies the choice of the rank sum test statistic. For example, list(name="Wilcoxon") means the Wilcoxon rank sum statistic, and list(name = "Stephenson", s = 10) means the Stephenson rank sum statistic with parameter s=10.
score	An n dimensional transformed ranks, i.e., (phi(1), phi(2), ..., phi(n)), where phi() denotes the rank transformation function.
stat.null	An vector whose empirical distribution approximates the randomization distribution of the rank sum statistic.
nperm	A positive integer representing the number of permutations for approximating the randomization distribution of the rank sum statistic.
calc_ICC	Calculate how much power varies across finite datasets.
quantile_n_CI	What quantile of the distribution of number of significant units should be reported?
...	Extra parameters to generate treatment impacts with.

Details

Each individual dataset is tested iter_per_set times, giving very rough power calculations for individual sets that are then aggregated; this provides the ability to do variance decomposition to see how sensitive power is to individual dataset characteristics (for datasets of the same family as defined by the DGP) vs. randomization imbalance.

li-xinran/RIQITE documentation built on July 1, 2023, 6:58 p.m.