fisher.power

This function calculates the power by simulations using Fisher's exact test.

Implements Bayesian hybrid designs that incorporate historical
control data into a current clinical trial. The package uses a dynamic
power prior method to determine the degree of borrowing from the historical
data, creating a 'hybrid' control arm. This approach is primarily
designed for studies with a binary primary endpoint, such as the overall
response rate (ORR). Functions are provided for design calibration,
sample size calculation, power evaluation, and final analysis.
Additionally, it includes functions adapted from the 'SAMprior' package
(v1.1.1) by Yang et al. (2023) <https://academic.oup.com/biometrics/article/79/4/2857/7587575> to support the
Self-Adapting Mixture (SAM) prior framework for comparison.

Zhaohua Lu

BayesianHybridDesign

Bayesian Hybrid Design and Analysis

Philip He

fisher.power function

<dl><dt>pt</dt>
<dd>A scalar. Probablity of success in experimental arm</dd>
<dt>nt</dt>
<dd>A scalar. Number of subject in experimental arm</dd>
<dt>pc</dt>
<dd>A scalar. Probablity of success in control arm</dd>
<dt>nc</dt>
<dd>A scalar. Number of subject in control arm</dd>
<dt>alpha</dt>
<dd>A scalar. One sided type I error rate.</dd>
<dt>nsim</dt>
<dd>A scalar. Number of replications to calculate power. Default
100,000.</dd>
<dt>seed</dt>
<dd>An integer. A seed for the random number generator to ensure
reproducibility. Note: For best practice, it is recommended to call
<code><a href="/link/set.seed()?package=BayesianHybridDesign&version=0.1.0" data-mini-rdoc="BayesianHybridDesign::set.seed()">set.seed()</a></code> outside of the function.</dd></dl>

Arguments

Power Calculation by Fisher's Exact Test — fisher.power

<dl>

<dt>pt</dt>
<dd>A scalar. Probablity of success in experimental arm</dd>


<dt>nt</dt>
<dd>A scalar. Number of subject in experimental arm</dd>


<dt>pc</dt>
<dd>A scalar. Probablity of success in control arm</dd>


<dt>nc</dt>
<dd>A scalar. Number of subject in control arm</dd>


<dt>alpha</dt>
<dd>A scalar. One sided type I error rate.</dd>


<dt>nsim</dt>
<dd>A scalar. Number of replications to calculate power. Default
100,000.</dd>


<dt>seed</dt>
<dd>An integer. A seed for the random number generator to ensure
reproducibility. Note: For best practice, it is recommended to call
<code><a href='https://rdrr.io/r/base/Random.html'>set.seed()</a></code> outside of the function.</dd>

</dl>

fisher.power: Power Calculation by Fisher's Exact Test

Description

Usage

Value

Arguments

Examples