f.learn.confirm: Main Function for Learn-Confirm Analysis

In gu-mi/simASD: Enhancement of the Adaptive Signature Design (ASD) for Learning and Confirming in a Single Pivotal Trial

Description

This function takes a mega dataset and returns test significances for one-stage design, two-stage design (all-comer and sensitive group). This is in preparation for function f.pwr.cal where multiple combinations are considered.

Usage

f.learn.confirm(data, learn.allocation = 30, cutoff = 0.4,
  ac.alpha = 0.04, sg.alpha = 0.01, verbose = FALSE)

Arguments

data	pass a mega dataset (including multiple individual datasets)
learn.allocation	percentage of all subjects allocated to the learn stage
ac.alpha	alpha level assigned to (two-stage) all-comer analysis
sg.alpha	alpha level assigned to (two-stage) sensitive group analysis
verbose	(logical) whether to print out results for each trial (default: FALSE)

Value

a list of three vectors for indicating trial results, and other quantities for the integrated power analyses performed by the function f.pwr.cal

Author(s)

Gu Mi <[e-mail]>

See Also

f.learn.1 for learn stage analysis for a single dataset f.confirm.1 for confirm stage analysis for a single dataset

gu-mi/simASD documentation built on May 17, 2019, 8:57 a.m.