Package: bbssr
Type: Package
Title: Blinded Sample Size Re-Estimation for Binary Endpoints
Version: 3.0.0
Authors@R: person("Gosuke", "Homma", email = "my.name.is.gosuke@gmail.com", 
                  role = c("aut", "cre"))
Description: Tools for blinded sample size re-estimation (BSSR) in
        two-arm clinical trials with binary endpoints, together with
        the exact power and sample size calculations that the
        re-estimation relies on, for two independent groups, for
        matched pairs and for stratified designs. For two independent
        groups seven statistical tests are implemented: Pearson
        chi-squared, Fisher exact, Fisher mid-p, Z-pooled exact
        unconditional, the exact unconditional test of Boschloo (1970)
        <doi:10.1111/j.1467-9574.1970.tb00104.x>, and the tests of
        non-inferiority and equivalence of Blackwelder (1982)
        <doi:10.1016/0197-2456(82)90024-1> and of Farrington and
        Manning (1990) <doi:10.1002/sim.4780091208>, with a margin on
        the scale of the risk difference or the risk ratio. Each test
        is available with a one-sided or a two-sided alternative, and
        the exact unconditional tests can be combined with the
        Berger-Boos procedure. For matched pairs the McNemar tests and
        the score test of Nam (1997) <doi:10.2307/2533508> are
        implemented, and for stratified designs the
        Cochran-Mantel-Haenszel test, Cochran's test with the sample
        size formulas of Woolson et al. (1986) <doi:10.2307/2530706>
        and Nam (1992) <doi:10.2307/2532298>, and the stratified exact
        conditional test of Jung (2014) <doi:10.1002/bimj.201300048>.
        The power, the distribution of the final sample size and the
        type I error rate of a re-estimation design are computed
        exactly, by simulation for three or more strata and as averages
        over simulated allocation sequences for random stratum and
        group sizes, the significance level can be adjusted so that the
        type I error rate is controlled, and the sample size can be
        re-estimated from the blinded data of a trial that is under
        way. Statistical methods based on Mehrotra et al. (2003)
        <doi:10.1111/1541-0420.00051>, Berger and Boos (1994)
        <doi:10.1080/01621459.1994.10476836>, Friede and Kieser (2004)
        <doi:10.1002/pst.140>, Friede et al. (2007)
        <doi:10.1002/bimj.200610373> and Kieser (2020)
        <doi:10.1007/978-3-030-49528-2_21>.
License: MIT + file LICENSE
URL: https://gosukehommaex.github.io/bbssr/,
        https://github.com/gosukehommaEX/bbssr
BugReports: https://github.com/gosukehommaEX/bbssr/issues
Depends: R (>= 3.5.0)
Imports: fpCompare, ggplot2 (>= 3.4.0), Rcpp, stats, utils
LinkingTo: Rcpp
Suggests: testthat (>= 3.1.7), knitr, rmarkdown, Exact, exact2x2,
        microbenchmark
VignetteBuilder: knitr
Encoding: UTF-8
Language: en-GB
Roxygen: list(markdown = TRUE)
Config/testthat/edition: 3
Config/roxygen2/version: 8.0.0
Repository: https://gosukehommaex.r-universe.dev
Date/Publication: 2026-10-28 00:26:39 UTC
RemoteUrl: https://github.com/gosukehommaex/bbssr
RemoteRef: HEAD
RemoteSha: 976931545a22d5628200e12f7584c3f1d7d8c875
NeedsCompilation: yes
Packaged: 2026-10-28 06:09:57 UTC; root
Author: Gosuke Homma [aut, cre]
Maintainer: Gosuke Homma <my.name.is.gosuke@gmail.com>