Lawrence Joseph
Bayesian statistics


Please Note: Prof. Joseph has retired.
These pages are left up in case they prove
useful, but the pages and software will
no longer be updated. All material and
software is "as is" with no guarantees
of functionality or correctness.
  • > Home
  • >
  • Bayesian Software
  • >
  • Bayesian Sample Size
Bayesian Sample Size
See related papers
Change-point methods and applications Diagnostic testing Diagnostic testing in Genetics
Please Note: Prof. Joseph has retired. These software packages remain available in case they prove
useful, but the software will no longer be updated or supported. All material and software is "as is"
with no guarantees of functionality or correctness.
CBSS
Consensus-Based Sample Size
Version 1.0, July 2019
Four sets of R functions for calculating sample size requirements to ensure posterior agreement from different priors using a variety of Bayesian criteria. Each package includes functions for designing an experiment to estimate, respectively:
  • a single binomial proportion
  • the difference between two binomial proportions
  • a single normal mean
  • the difference between two normal means.
This package is an implementation of the methods presented in
Bayesian consensus-based sample size criteria for binomial proportions
Lawrence Joseph and Patrick Bélisle
Statistics in Medicine; 2019
Zipped R packages
SampleSizeConsensusBinomialProportion_1.0.tar.gz
SampleSizeConsensusBinomialProportionsDiff_1.0.tar.gz
SampleSizeConsensusNormalMean_1.0.tar.gz
SampleSizeConsensusNormalMeansDiff_1.0.tar.gz
Install instructions
For each of the four packages, follow the instructions written for SampleSizeBinomial.
For both BinomialProportion packages, follow each step listed in the instructions, as these packages include a .C program.
For the other two packages (NormalMean), skip the first two steps.
SampleSizeRegression
Bayesian Sample Size Criteria for Linear and Logistic Regression in the Presence of Confounding and Measurement Error
Version 1.0, July 2019
A package to calculate Bayesian sample sizes to estimate linear or logistic regression model parameters, with or without the possibility of confounding variables and/or independent variable measurement error; requires that the free software packages R, Winbugs and Perl be installed.
This package is an implementation of the methods presented in
Instructions
InstructionManual.pdf
One-click download
Install instructions
  1. download and unzip the zipped file (above)
  2. double-click setup.exe
  3. follow the instructions in file doc\InstallInstructions.html
    (or in its online version)
SSCOR
Odds Ratio in presence of exposure misclassification
Version 1.8.1, February 2019
A package to calculate sample sizes based on highest posterior density intervals in the context of estimating odds ratios when exposure may be misclassified; requires that the free software packages R, Winbugs and Perl be installed.
This package is an implementation of the methods presented in
Bayesian sample size determination for case-control studies when exposure may be misclassified
Lawrence Joseph and Patrick Bélisle
American Journal of Epidemiology
Accepted
Instructions
InstructionManual.pdf
One-click download
Install instructions
  1. download and unzip the zipped file (above)
  2. double-click setup.exe
  3. follow the instructions in file doc\InstallInstructions.html
    (or in its online version)
PropMisclassSampleSize
Prevalence and diagnostic test sensitivity or specificity
Version 6.0, February 2019
A package to calculate sample sizes based on highest posterior density intervals in the context of diagnostic testing in presence of one, two or three imperfect tests; requires that the free software packages R, Winbugs and Perl be installed.
This package is an implementation of the methods presented in
Bayesian Sample Size Determination for Prevalence and Diagnostic Test Studies in the Absence of a Gold Standard Test
Nandini Dendukuri, Elham Rahme, Patrick Bélisle and Lawrence Joseph
Biometrics 60, pp. 388-397
June 2004
Instructions
InstructionManual.pdf
One-click download
Install instructions
  1. download and unzip the zipped file (above)
  2. double-click setup.exe
  3. follow the instructions in file doc\InstallInstructions.html
    (or in its online version)
SampleSizeMeans
Normal means
Version 1.0, December 2009
A set of R functions for calculating sample size requirements using three different Bayesian criteria in the context of designing an experiment to estimate a normal mean or the difference between two normal means.
This package is an implementation of the methods presented in
Available from CRAN
SampleSizeMeans R package
SampleSizeBinomial
Binomial proportion
Version 1.1, March 2018
A set of R functions for calculating sample size requirements using three different Bayesian criteria in the context of a binomial experiment.
This package is an implementation of the methods presented in
Sample size calculations for binomial proportions via highest posterior density intervals
Lawrence Joseph, David B. Wolfson and Roxane du Berger
The Statistician, 1995;44(2):143-154
Zipped R package
SampleSizeBinomial_1.1.tar.gz
Install Instructions
Install Instructions
bhpd1
Binomial proportion
A FORTRAN program to calculate sample sizes based on highest posterior density intervals in the context of a binomial experiment using three different Bayesian approaches.
This package is an implementation of the methods presented in
Sample size calculations for binomial proportions via highest posterior density intervals
Lawrence Joseph, David B. Wolfson and Roxane du Berger
The Statistician, 1995;44(2):143-154
Instructions
Instructions
Ex.input
Files
Makefile
bhpd.f
cover.f
headers.f
search.f
baysmp.f
frqsmp.f
readpar.f
util.f
One-click download
Binomial proportion in the presence of misclassification error
An S-Plus program to calculate average coverage probabilities when finding sample sizes for estimating a binomial proportion in the presence of misclassification errors.
This package is an implementation of the methods presented in
Instructions
Instructions
Files
avgcov-ss.txt
One-click download
SampleSizeProportions
Binomial proportions
Version 1.0, December 2009
A set of R functions for calculating sample size requirements using three different Bayesian criteria in the context of designing an experiment to estimate the difference between two binomial proportions.
This package is an implementation of the methods presented in
Bayesian and mixed Bayesian/likelihood criteria for sample size determination
Joseph L, du Berger R, and Belisle P
Statistics in Medicine 1997;16(7):769-781
Available from CRAN
SampleSizeProportions R package