PAC
Programme for the Analysis of Comparisons
FORTRAN Source 		LePAC
WINDOWS software 		LePrep
WINDOWS software
WORD macro
Statistical distributions
for  the Bayesian analysis ofexperimental data
FORTRAN functions 		Blaker's confidence interval
FORTRAN source
R Source


All programs presented here are free and open source software

The users have the freedom to run, copy, distribute, study, change and improve the software.

We defend these freedoms because everyone deserves them. With these freedoms, the users (both individually and collectively) control the program and what it does for them. In contrast the nonfree program controls the users, and the developer controls the program; this makes the program an instrument of unjust power.

What is free software?

Authors

Bruno Lecoutre Bruno LECOUTRE   Jacques Poitevineau Jacques POITEVINEAU


PAC
Programme for the Analysis of Comparisons
FORTRAN Source 		LePAC
WINDOWS software 		LePrep
WINDOWS software
WORD macro
Statistical distributions
for  the Bayesian analysis ofexperimental data
FORTRAN functions 		Blaker's confidence interval
FORTRAN source
R Source



PAC (Programme for the Analysis of Comparisons) is a general univariate and multivariate variance analysis program. It is a unique tool, usable as well for an initiation to data processing, for both experimental and sophisticated applications. It is a powerful general program of univariate variance analysis, and multivariate, including covariance analysis, polynomial regression analysis and discriminant analysis.

It is particularly suitable for repeated measurement designs, with equal or uneven group sizes.

A powerful language for comparisons allows the user, depending on his questions, to select the effects to analyze: main, interaction, partial, conditional, within, associated with polynomial regression components.

PAC performs univariate and multivariate, descriptive and quantitative analyses, and inferential, detailed. These analyses include the following procedures and extend them, on the one hand, by making it possible to abandon the usual assumptions about variance-covariances, and on the other hand, by providing conclusions about the true magnitude of the effects (especially in the context of Bayesian inference).
PAC was designed in the 1980s by Jacques Poitevineau, in collaboration with Bruno Lecoutre for the statistical part itself. It was written by Jacques Poitevineau in the FORTRAN 90 standard. The subroutines are original, with the exception of a few from libraries. LINPACK/EISPACK, and some mathematical functions. like betacdf. f90 (when this is the case, the origin is indicated by comments in the source).

Originally, PAC was designed to run alone on most platforms in console mode.

PAC consists of two main subsets:
1) The GMP subset (for Groupe Mathématiques et Psychologie). The purpose of the subprograms in this subset is the general processing of data (how many variables, their characteristics), What are the names to be given to variables, the name of the input data file, etc.?
2) The PAC subset that performs statistical analyses

Each of these subsets uses the ANACOM7 toolbox (ANAlyse des COMmandes) for the analysis of user commands. The structure and syntax of these commands, as well as some aspects of the program's organization, were inspired by BMDP software.


PAC
Programme for the Analysis of Comparisons
FORTRAN Source 		LePAC
WINDOWS software 		LePrep
WINDOWS software
WORD macro
Statistical distributions
for  the Bayesian analysis ofexperimental data
FORTRAN functions 		Blaker's confidence interval
FORTRAN source
R Source


Le P A C   Version 2.0
LePAC was originally a user-friendly Windows interface for running PAC.

Other programmes have been gradually included:

LesMoyennes implements in a simple and interactive way Bayesian procedures for average comparison problems. It includes the usual frequent dentist procedures, tests of significance and confidence intervals. It also makes accurate inferences for a correlation coefficient: Tests (null hypothesis ρ=0 ou ρ=ρ0≠0), confidence intervals, fiducial-Bayesian procedures.

LesEffectifs calculates Bayesian predictive probabilities and incorporates the power of traditional tests as a special case.

LesEchantillons runs sample drawings and simulations of sampling distributions for descriptive and inferential statistics in the case of one or two samples of normal parent distributions.

LesProportions makes Bayesian inferences on proportions.

LeB-A-Bayésien has a didactic vocation to understand the B-A-BA. Bayesian inference from basic situation.

LesDistributions allows to obtain both density curves, direct and inverse distribution functions, random (including cumulated) functions, and probability statements for direct and inverse distribution functions. A large number of continuous and discrete distributions are available, with transformations for continuous distributions and the possibility of lower (left) truncation and upper (right) truncation.

LePrep can be used from a student t test or a F analysis of variance test with one degree of freedom to obtain the predictive probability prep ("Killeen's probability of replication"). to find an effect of the same sign in a replication.
LePAC was written in Visual Basic by Bruno Lecoutre. Jacques Poitevineau wrote many functions in FORTRAN.
To know more
about LePAC

PAC
Programme for the Analysis of Comparisons
FORTRAN Source 		LePAC
WINDOWS software 		LePrep
WINDOWS software
WORD macro
Statistical distributions
for  the Bayesian analysis ofexperimental data
FORTRAN functions 		Blaker's confidence interval
FORTRAN source
R Source


LePrep "Probabilities of replication"   

LePrep is an independent version of the program included in LePAC. It can also be used as a Word macro.
Computation of Killen's prep, psrep and preprep
Interval estimation for contrast: standardized (calibrated) or raw effect
K-prime distribution (special cases: t non-central and Lambda-prime)

 Lecoutre B., Lecoutre M.-P., Poitevineau J. (2010) - Killeen’s probability of replication and predictive probabilities: How to compute, use and interpret them. Psychological Methods, 15, 158-171.

To find out more
about LePrep

PAC
Programme for the Analysis of Comparisons
FORTRAN Source 		LePAC
WINDOWS software 		LePrep
WINDOWS software
WORD macro
Statistical distributions
for  the Bayesian analysis ofexperimental data
FORTRAN functions 		Blaker's confidence interval
FORTRAN source
R Source


Statistical distributions
                   pour l'analyse bayésienne des données expérimentales
                   Fonctions FORTRAN

Programs in FORTRAN are more specifically devoted to statistical distributions for the Bayesian inference. In particular, they give the distribution functions of distributions introduced for the Bayesian analysis of comparisons: Psi-square, Lambda-prime, Lambda-square, K-prime and K-square.

Lecoutre B., Guigues J.-L., Poitevineau J. (1992) - Distribution of quadratic forms of multivariate Student variables. Applied Statistics, 41, 617-627.
Lecoutre B. (1999) - Two useful distributions for Bayesian predictive procedures under normal models. Journal of Statistical Planning and Inference, 77, 93-105.
Poitevineau J., Lecoutre B. (2010) - Implementing Bayesian predictive procedures: The K-prime and K-square distributions. Computational Statistics & Data Analysis, 54, 723-730.

Download
Poitevineau J., Lecoutre B. (2010) - Statistical distributions for Bayesian experimental data analysis, FORTRAN functions: 1. Continuous distributions. ERIS.
Poitevineau J., Lecoutre B. (2010) - Statistical distributions for Bayesian experimental data analysis, FORTRAN functions: 2. Discrete distributions. ERIS.
Poitevineau J., Lecoutre B. (2010) - Statistical distributions for Bayesian experimental data analysis, FORTRAN functions: 3. Miscelleanous. ERIS.
Download
FORTRAN files (ZIP file)

PAC
Programme for the Analysis of Comparisons
FORTRAN Source 		LePAC
WINDOWS software 		LePrep
WINDOWS software
WORD macro
Statistical distributions
for  the Bayesian analysis ofexperimental data
FORTRAN functions 		Blaker's confidence interval
FORTRAN source
R Source


Blaker's exact confidence interval

Programs in FORTRAN and R calculate the exact Blaker confidence interval. for a proportion for a parent distribution:
Binomial
Negative binomial
Poisson
Hypergeometric

 Lecoutre B., Poitevineau J. (2016) - New Results for Computing Blaker's Exact Confidence Interval for One Parameter Discrete Distributions. Communications in Statistics Simulation and Computation, 45, 1041-1053.

Download
FORTRAN source (ZIP file)
Download
R source (ZIP file)


PAC
Programme for the Analysis of Comparisons
FORTRAN Source 		LePAC
WINDOWS software 		LePrep
WINDOWS software
WORD macro
Statistical distributions
for  the Bayesian analysis ofexperimental data
FORTRAN functions 		Blaker's confidence interval
FORTRAN source
R Source