AnalyzePerformanceSimon: Performance of estimation methods

In arsenionhacolo/InferenceBEAGSD: Inference for Binary Endpoint Adaptive Group-Sequential Designs

Description

It takes the results produced by AnalyzeSimonDsgn and AnalyzeSimonDsgnAdaptN and produces a dataframe containing bias, mean square error and variance of the estimators. It also calculates the power and the expected sample size (EN) where applicable.

Usage

AnalyzePerformanceSimon(designs = "all", basedir = NA)

Arguments

designs	Taking values "fixed", "adaptive" or "all", indicating whether only classical, adaptive or all designs should be included. The default is "all".
basedir	The root directory in which simulations were performed. The current working directory is assumed by default. It must contain all the files and folders created by SimulateSimonDsgn and/or SimulateSimonDsgnAdaptN.

Details

Computations are done for different combinations of values of stop, (0,1), and success, (0,1). See AnalyzeSimonDsgn or AnalyzeSimonDsgnAdaptN. For instance, computations done on all simulated trials are marked with "both" in the columns stop and success, while the ones done only on trials that continued to the final stage have stop = "no" and success = "both".

Value

Dataframe containing bias, mean square error and variance of the estimators, power, expected sample size, and design information.

Author(s)

Arsenio Nhacolo

See Also

AnalyzeSimonDsgn, AnalyzeSimonDsgnAdaptN, pdata and AnalyzePerformanceSimon2.

Examples

## Not run: 
AnalyzePerformanceSimon()

## End(Not run)

arsenionhacolo/InferenceBEAGSD documentation built on May 9, 2019, 4:10 a.m.