StatLib---General Archive
This collection contains a variety of software written in Fortran, C and Lisp, some complete statistical systems
and other odds and ends.
To contribute software to this archive, please obtain a copy of the submission instructions: send mail to statlib
and ask send submissions from general The software is not warranteed in any way. Unless otherwise stated the
software is exactly what is provided by the submitters. All questions and comments should be directed to the
submitter.
Some of the entries are shar archives. If you don't know how to deal with a shar archive, send the message send
shar from generalfor instructions.
------------------------------------------------------------------------------------------------------------------
The general archive currently contains:
submissions
The instructions for how to submit software to statlib. [1/May/89] (1673 bytes)
accflf
See utexas below. Program to fit very general accelerated failure model with log-F error to possibly
censored data.
ace
Brieman and Friedman's ACE algorithm (in fortran). Estimates optimal transformations for multiple
regression [10/Oct/89][5/Jan/94] (38 kbytes)
adj_ptsl.tar.gz
SAS code which estimates the median survival of the standard therapy database adjusted for the patient
distribution observed in the phase II study using the booststrap resampling technique. A gzipped tar
archive. Submitted by David Verbel (verbeld@biost.mskcc.org) [25/Feb/00] (2394 bytes)
ajay
C functions for cholesky decomposition, and multivariate normal generation. (Ajay Shah,
ajayshah@rand.org) [23/Sept/91] (7194 bytes)
analyst
Fortran programs and datasets from Analyst 114, 1693-1702 (1989) for robust statistics in analytical
chemistry. Submitted by Brian Ripley (ripley@stats.ox.ac.uk) [31/Mar/91](24419 bytes)
assoc
Fortran program using the scoring algorithm for association analysis described in the Gilula and
Haberman 1986 paper, "Canonical Analysis of Contingency Tables by Maximum Likelihood," in Journal of the
American Statistical Association, 81, 780-788. See also dassoc. Shelby Haberman (shelby@fisher.stats.nwu.edu)
[17/Mar/94] (24 k)
autologit
C++ code for the implementation of the autologistic model with covariates. A shar archive. Submitted by
Greg Young (young@stat.colostate.edu) [9/Apr/01] (393 Kbytes)
baysyn
Bayesian synthesis method for bowhead whale population dynamics model (Raftery, Givens, Zeh 1995
JASA). Fortran. Also see bergs. Submitted by Geof Givens (geof@lamar.ColoState.EDU) [1/Mar/95] (266 kbytes)
baysub
This SAS program implements Bayesian Subset Analysis according to "Bayesian Subset Analysis" D O Dixon
and R Simon (Biometrics 91) Submitted by BORIS_FREIDLIN_at_EMMES@emmes.com [28/Aug/95](18 kbytes)
bcv.sas
SAS Macro Subroutine in order to Calculate the Akaikes Information Criterion (AIC), Mallows Cp, PRESS
(Prediction Sum of Squares), an approximation for PRESS, a Bootstrap-based extension to AIC (called EIC), a
Bootstrap-smoothed likelihood cross-validation (BCV) and its modification (632CV). Submitted by Paul Johnson
(p.johnson@prodigy.net). [21/Jan/00][2/Jan/01] (10 kbytes)
bergs
Reweighting sensitivity analysis for Bayesian synthesis bowhead whale pop. assessment (Raftery, Givens,
Zeh 1995 JASA). Fortran. See baysyn. Submitted by Geof Givens (geof@stat.ColoState.EDU) Available only via
WWW or email. [1/Mar/95] (2.46 Mbytes)
bhpd1
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. Submitted by Lawrence Joseph
(joseph@binky.epi.mcgill.ca), David Wolfson and Roxane du Berger. [23/Jan/95](107k)
bioq.sas
SAS Macro: Modified large sample upper bound for the Bioequivalence Ratio. Submitted by Paul Johnson
(P.Johnson@prodigy.net) [2/Jan/01] (8k)
bkd
A link to Bayesian Knowledge Discoverer, a program designed to extract Bayesian Belief Networks (BBNs)
from (possibly incomplete) databases. It is based on a new estimation method called Bound and Collapse and
its extensions to model selection. Paola Sebastiani (P.Sebastiani@city.ac.uk)
bmds
FORTRAN program to implement the methods described in "Bayesian Multidimensional Scaling and Choice of
Dimension" by M-S. Oh and A. Raftery, JASA 2001, vol. 96. Submitted by Man-Suk Oh
(msoh@mm.ewha.ac.kr). [27/Aug/01] (11332 bytes)
bootcox
SAS Macro Using the Bootstrap to Estimate the Survival Function (Cox Regression). Paul Johnson
(EZ006244@EUCLID.UCDVIS.EDU) [5/Jun/96] (6k)
bootkm
SAS Macro for the Kaplan-Meier Bootstrap Estimate of the Survival Function Paul Johnson
(EZ006244@EUCLID.UCDVIS.EDU) [5/Jun/96] (6k)
bootlogit
SAS Macro to Calculate Bootstrap Estimates of Probabilities in Logistic Regression Paul Johnson
(EZ006244@EUCLID.UCDVIS.EDU) [2/Aug/96] (6k)
bootnet
A Unix shell script for distributing bootstrap computations across a network. Arny Stromberg
(astro11@ukcc.uky.edu) [1/Jun/95] (6 kbytes)
Bmw.sas
SAS Macro: The Breiman and Meisal Assessment of Lack of Fit in Linear Regression. Submitted by Paul
Johnson (P.Johnson@prodigy.net) [22/Feb/00][2/Jan/01] (7 kbytes)
bvls
A FORTRAN program to solve linear least-squares problems with upper and lower bounds on the variables,
using an active set strategy. It is documented in the J. of Computational Statistics, and can be used
iteratively to solve minimum l-1, l-2 and l-infinity fitting problems. Robert L. Parker and Philip B. Stark,
(stark@gandalf.Berkeley.EDU) [1/Mar/95] (20k)
bystub
A FORTRAN program for Bayesian analysis of tube dilution data Published in B.Roussanov, D.M.Hawkins and
S.R.Tatini (1996). Estimating Bacterial Density from Tube Dilution Data by a Bayesian Method. Food
Microbiology 13, 341-363. Submitted by Doug Hawkins, doug@stat.umn.edu [28/Feb/96] (8k)
ccount
A package for market research data cleaning, manipulation, cross tabulation and data analysis. Similar
to, and uses the same syntax as, SPSS-MR "Quantum", a well known commercial package for processing market
research data. Submitted by Volker Hoffmann (vhoff@pan-data.dyndns.org) [14/Oct/03] (237k)
cdflib
Fortran routines cumulative distribution functions, their inverses, and their parameters for Beta,
Binomial, Noncentral Chi-square, Noncentral F, Gamma, Negative Binomial, Normal, Poisson, Student's t. (Barry
Brown, bwb@odin.mda.uth.tmc.edu) [20/Aug/93]
cdflib.shar
Shar file for cdflib. Its big (360k). Please FTP or WWW if possible.
cdh
Tests of composite distributional hypotheses for the analysis of biological and environmental data This
library contains FORTRAN subroutines for testing the hypothesis of normality or the hypothesis of
exponentiality. Paul Johnson (EZ006244@alcor.ucdavis.edu) [11/Mar/94] (48k)
cis94
This is the latest version of usrguide.txt, the ASCII text form of the User Guide to the 1994 Edition of
the Current Index of Statistics Extended Database. It is the same as the version distributed in the DOS and
MAC directories of the CD-ROM, but supersedes the version distributed in the UNIX directory of the CD-ROM and
on the diskettes. It differs from the older version by being better suited for viewing at the terminal and by
having appendices describing the abbreviation files and the subject index. Michael Wichura
(wichura@galton.uchicago.edu). [23/Jan/95] (161 k)
cissearch
A shar archive of software for using the on-line data base for the Current Index to Statistics. Updated
for the 1994 version. (see also cis-software in the asascs collection). Submitted by Michael Wichura
(wichura@galton.uchicago.edu). [4/Sept/92] [23/Jan/95](245751 bytes)
clusfind
A set of six stand-alone Fortran programs for cluster analysis. The programs are described and
illustrated in the book "Finding Groups in Data" by L. Kaufman and P.J. Rousseeuw, New York: John
Wiley. Includes all data sets from the book. Submitted by Peter Rousseeuw
(rousse@wins.uia.ac.be). [23/Jan/95] (254 kbytes)
concor
CONCOR method analyses the links of a set with several sets (concor, concoreg,concorcano) or analyses
the links of a multiset with another multiset (concorgm,concorgmreg,concorgmcano). For calculating a set of
simultaneous solutions another algorithm is necessary (concors, concorsreg, concorscano). This method is
based on a SVD notion extended to a partitioned matrix (svdcp, svdbip or svdbip2, and svdbips). The
associated papers run from 1997 to today, with the following authors. Cazes Pierre, Ceremade, Univ. Dauphine,
Paris 9, France Kissita Gabriel, Univ. Marien N'Gouagi, Brazzaville, Congo Hanafi Mohamed, ENITIAA, Nantes,
France Lafosse Roger, Univ. Sabatier, Toulouse, France Ten Berge, Heymans Inst., Groningen, Netherlands The
functions written in R langage may be loaded by source("concorR"). [6/Jul/04] (7.3kbytes)
conhull
Planar convex hull code in Fortran. Submitted by Bill Eddy (bill@stat.cmu.edu) [4/May/92] (11584 bytes).
corbin
"A Fortran 77 function to generate correlated binary data (using the algorithm of Ahn & Chen). Requires
IMSL. Notes in postscript format. Submitted by Dario Gregori (gredar@univ.trieste.it), Loredana DiConsiglio,
Gaetano Carmeci" [2/Oct/95] (126k)
cmsbd
Collection of S functions with interface to Fortran and C, that calculate properties of continuously
monitored stopping boundaries for clinical trials. A compressed tar file. Submitted by Emmanuel Lazaridis
(lazarien@moffitt.usf.edu). For more information see http://www.moffitt.usf.edu/biostat/lazaridis/design
[3/Jun/98] (354 kbytes)
count
A pointer to a stand-alone DOS program for running event count and duration regression models, developed
by and/or discussed in a series of journal articles by Gary King. [4/Jan/93]
cutoff
A FORTRAN program for establishing optimal cutpoints for screening and diagnostic tests, including
documentation. Submitted by Dean McKenzie (MPY148Y@vaxc.cc.monash.edu.au) [5/Feb/1993] (13kb)
dassoc
Fortran program using the modified Newton-Raphson algorithm for association models described in the
Haberman 1994 paper, "Computation of Maximum-Likelihood Estimates in Association Models." This paper has been
submitted for publication. See also assoc. Shelby Haberman (shelby@fisher.stats.nwu.edu) [17/Mar/94] (25 k)
days85.for
Interactive FORTRAN program for case-control design sample size required to detect a given odds-ratio
value. Same as daysmith.for except for 85% and 95% power levels. Contact author for alternative power levels
Windows executable in days85.exe. Submitted by Michael Edwardes (michael.edwardes@clinepi.mcgill.ca)
[2/Nov/01][4/Dec/01] (8k)
daysmith.for
Interactive FORTRAN program for case-control design sample size required to detect a given odds-ratio
value at 80% and 90% power. Adjusts for a given confounder, and for effect modification . Documentation
currently at http://www.biomedcentral/1471-2288. Windows executable in daysmith.exe. See
design.for. Submitted by Michael Edwardes (michael.edwardes@clinepi.mcgill.ca)
[3/Jun/98][23/Jul/01][2/Nov/01][4/Dec/01] (8k)
delaunay
A suite of ratfor subroutines and a driver (which must be edited to conform to a particular data set)
with which to calculate the Dirichlet or Voronoi tesselation and the Delaunay triangulation of a planar point
set. [13/Dec/96] (67k) See also the S version delaunay in the S collection.
design.for
Interactive FORTRAN program for case-control design sample size required to detect a given odds-ratio
value. Adjusts for a given confounder. daysmith.for does the same task with different input. Documentation
currently at http://www.biomedcentral/1471-2288. Windows executable in design.exe Submitted by Michael
Edwardes (michael.edwardes@clinepi.mcgill.ca) [3/Jun/98] [23/Jul/01][2/Nov/01][4/Dec/01](10k)
designv1
A FORTRAN program for finding locally and Bayesian optimal designs for binary dose-response models with
various link functions. Submitted by David Smith (d.smith@icrf.icnet.uk)
[4/Dec/98][3/Feb/99][2/Jun/99][24/May/00] (137k)
designv2
A FORTRAN program for calculating the optimal designs for the log(potency) of comparative binary
dose-response models with no control mortality and alternative link functions. It also evaluates these
designs using General Equivalence Theorem results. A related article is to be published in the Journal of
Statistical Planning & Inference in 2002/3. Submitted by David Smith (davismith@mail.mcg.edu) [1/Nov/02]
(176k)
distrib
An addin package for ASEASYAS, a LOTUS clone spreadsheets. It calculates values for central and
noncentral distribution functions, and can be used for calculating the power of standard tests
easily. Uuencoded, zip format file. Submitted by Erich Neuwirth (A4422DAB@AWIUNI11.BITNET) [19/Nov/91] (29301
bytes)
dysect
Clustering algorithm based on dynamic altering of hierarchies. Suitable for large data sets. Submitted
by (oconnor@isor.vuw.ac.nz) [19/Sep/94] ( 438k)
dyn.dec
Dynamic programming and related programs for Bayesian Decision Analysis used in: "Decision Making During
a Phase III Randomized Controlled Trial", CONT CLIN TR. Mark C. Wolff
(mcw2c@brain.neuro.virginia.edu). [20/Dec/93] (15 k)
eff_var
SAS Macro: Testing whether variances differ and, if they differ, the form of the difference. Submitted
by Paul Johnson (P.Johnson@prodigy.net) [21/Jul/99][2/Jan/01] (9k)
eha
Discrete time event history analysis, written for World Fertility Survey data, but nearly all of the
subroutines are more generally applicable. Steven M. Lewis (slewis@stat.washington.edu) [1/Mar/95] (347 k)
elicit-normlin
This program is designed to elicit the hyperparameters of the prior distribution for the normal linear
model, based on the 1980 JASA paper by Kadane et. al. Submitted by Lara Wolfson
(lara@stat.cmu.edu). [8/Aug/94][3/Jun/98] (60 k)
elicit-diric
Fortran program for Elicitation of concentration parameter of Dirichlet Process Priors Submitted by Ram
Gopalan (rmg@hooked.net) [13/Dec/96] (5k)
em_covar
A SAS/IML macro for estimating a covariance matrix and mean vector via the expectation maximization (EM)
algorithm. Bootstrap estimates are also available. Submitted by Steve Gregorich (Gregorich@psg.ucsf.edu)
[5/May/99] (42k)
esc.errata
Corrections and clarifications to Elements of Statistical Computing: Numerical Computation, by Ronald
Thisted (thisted@galton.uchicago.edu) (34 kbytes)[12/Nov/89][28/Jun/93]
ESS
Emacs Speaks Statistics(ESS): A multi-platform, multi-package intelligent environment for statistical
analysis. ESS extends the emacs text editing system (both GNU Emacs and XEmacs are supported). ESS
streamlines statistical analysis and the creation of statistical software with interfaces to packages like
S-PLUS, R, SAS, BUGS, Stata and XLispStat under the Unix, Microsoft Windows, and Apple Mac operating
systems. Mirror of http://ess.r-project.org/downloads/ess/ [17/Feb/02]
exact
FORTRAN program for exact, unconditional tests for 2X2 binomial and multinomial contingency tables. 1-
and 2-sided tests and three test statistics are supported. Roger L. Berger (berger@stat.ncsu.edu) (54k)
[13/Dec/96]
extlms
A set of Fortran subroutines and calling programs for computing the exact least median of squares
regression estimator under either a serial or distributed computing architecture. Submitted by Doug Hawkins
(doug@stat.umn.edu), Jeff Simonoff (jsimonoff@stern.nyu.edu) and Arny Stromberg
(astro11@ukcc.uky.edu). [20/Dec/93] [25/Jul/94] (66 kbytes)
fact
The source code for Wei-Yin Loh's FACT, classification and regression program. A stand alone
program. The interface to S is no longer supported. UUencoded, compressed, tar file. [10/Oct/89] (241338
bytes)
fls1.tar.gz
The dataset contains data on catches per species for a number of vessels collected by the EU in the
context of the NAFO. The f77 code implements the MC^3 algorithm. [C.Fernandez, E.Ley and M.F.J. Steel (1998)
"Bayesian Modeling of Catch in a NW Atlantic Fishery']. Gzipped tar file. Submitted by Eduardo Ley
(ELEY@IMF.ORG). [22/Jul/98] (128433 bytes)
fracdiff
Maximum likelihood estimation of the parameters of a fractionally-differenced ARIMA (p,d,q) model
(Haslett and Raftery, Applied Statistics 1989). Fortran source code in a shar archive. See also fracdiff in
the S collection of StatLib. Submitted by Chris Fraley, (fraley@stat.washington.edu) [7/Oct/91]
[24/Mar/91]. [4/Nov/91] [15/Sept/92] [22/Feb/93] [11/Mar/94] [4/May/94](220 kbytes)
fsa
A package of 5 FORTRAN routines using the feasible solution algorithms for high breakdown regression and
multivariate location/scatter estimation. "Douglas M. Hawkins" (doug@umnstat.stat.umn.edu) [22/Dec/93]
[17/Oct/94] (140 kbytes)
gamfit
FORTRAN program for fitting generalized additive models. Written by T. Hastie and
R. Tibshirani. Submitted by R. Tibshirani (tibs@utstat.toronto.edu). [1/Jan/91] [14/July/91] (280221 bytes)
gaut.c
C versions of routines to calculate tail probabilities for t, gaussian, and bivariate gaussian
distributions. Submitted by Ajay Shah (ajayshah@usc.edu) [12/May/91] (7848 bytes).
GeneticEpi
A link to the Human Genetic Analysis Resource, an NIH supported research project. From this research, we
have developed the Statistical Analysis for Genetic Epidemiology (S.A.G.E.) software package. Submitted by
Paul Gerwe (gerwe@darwin.mhmc.cwru.edu) [14/Oct/97]
gibbsit
Determines the number of iterations required to achieve a specified accuracy level in the Gibbs
sampler. A Fortran program. (Adrian Raftery, raftery@stat.washington.edu) [13/Sept/91] [1/Mar/95] (114
kbytes)
gkslib
A toolbox of statistical routines for MATLAB, including ordinal logistic regression, Poisson regression,
nonlinear regression with sums of exponentials, probability distributions, quadrature, matrix exponentials
and some other associated special functions and matrix operations. A tar file. Submitted by Gordon K Smyth,
(gks@maths.uq.oz.au) [1/Dec/95] [3/Jun/98] [7/Jun/99](112 kbytes)
GMTD
Software for computing the Mixture Transition Distribution model (MTD) for the modeling of high-order
Markov chains. Submitted by Andre Berchtold (Andre.Berchtold@themes.unige.ch). Also available in a PC
version.[1/Jun/99]
GOF.zip
SAS macro (and sample SAS dataset) to perform goodness-of-fit test for logistic regression model as
published in Pulkstenis E, Robinson T 'Two goodness-of-fit tests for logistic regression with continuous
covariates' Statistics in Medicine 2002; 21:79-93. bodytext: This software can be freely used for
non-commercial purposes and can be freely distributed. The author is willing to provide assistance where
required. Submitted by Erik_Pulkstenis@hgsi.com. [23/Apr/03] (11kbytes)
GOF_ordinal.zip
This program implements goodness-of-fit procedures for ordinal response regression modeling
corresponding to the following Reference: Pulkstenis E, Robinson T. Goodness-of-fit tests for ordinal
response regression models. Statistics in Medicine 2004; 23:999-1014. The author is willing to provide
assistance where required. Submitted by erik_pulkstenis@hgsi.com [23/Mar/04] (27kbytes)
gpower
(WWW ONLY, material is archived in Germany) GPOWER is a interactive, menu-driven Pascal-program for
IBM-compatible PC's and Apple Macintosh computers. It performs high-precision statistical power analyses for
the most common statistical tests in behavioral research, that is, t-, F-, and chi-square-tests. [28/Aug/95]
grkpack
A shar archive of Ratfor/Fortran routines for calculating smoothing splines ANOVA estimates of data from
exponential families. It is an extension of the rkpk. LaTeX document and examples are
included. (yuedong@umich.edu) (720421 bytes) [1/Mar/95]
hcfb
Average-linkage hierarchical cluster analysis program. Very easy to use. Better tree diagrams than
SAS. QuickBASIC and FORTRAN versions. John Uebersax (uebersax@phs.bgsm.wfu.edu) [11/Mar/94] (33k).
hmodel
An X tool that manipulates a collection of hierarchical models. Given a model optimality criterion
threshold it finds all models that are at least as good. Current implementation is limited to log-linear
models for categorical data and includes the Akaike and Bayes information model optimality
criteria. Submitted by George Ostrouchov (ost@azure.EPM.ORNL.GOV) [3/Dec/91] (175872 bytes)
ibmod
IBMOD applies the internally balanced state space model reduction algorithm of Crabbe & Young,
J. Stat. Comp. Simul., 1989 and 1991. Submitted by Brian Young (young@agsm.ucla.edu) [1/Oct/91] (239185
bytes).
judgeit
A pointer to a stand-alone DOS program for analyzing electoral systems and redistricting plans. This
implements numerous statistical procedures described in a series of articles by Andrew Gelman and Gary
King. [4/Jan/93]
kappa
Fortran program to computes the Kappa statistic of interrater reliability for categorical
judgments. Submitted by David Zubrow (zubrow@andrew.cmu.edu) [30/July/91] (11101 bytes)
kappa2
A SAS Macro for the kappa measure of agreement between two observers. Submitted by Andrzej Galecki
(agalecki@sph.umich.edu) [17/Oct/94] (11 kb)
latent2lv
Fortran program to provide the necessary sample statistics, including intra-class correlations for the
analysis of multilevel data described in "Latent variable modeling of longitudinal and multilevel data
(Muthen, 97 Sociological Methodology)" [2/Jun/98] Submitted by Booil Jo (booil@alpha.oac.ucla.edu)
laplacpg
Fortran program to calculate the Laplace approximation to the marginal likelihood for the Poisson-Gamma
(negative binomial) model. Susan Rosenkranz (slr@biostat.washington.edu) [29/Mar/94] (252 kbytes)
linalg
A package of Java translations of high quality public domain linear algebra routines --- many of the
BLAS (basic linear algebra subroutines) and the LINPACK Cholesky decomposition (dpofa, dposl, and dpodi), QR
decomposition (dqrdc and dqrsl), singular value decomposition (dsvdc), and LU decomposition (dgefa, dgesl,
and dgedi) routines. Further documentation and explanation of the routines can be found at
http://ws13.fpl.fs.fed.us/linear_algebra.html Submitted by Steve Verrill (steve@ws10.fpl.fs.fed.us). Also
available as a self-extracting zip file here. [10/Jan/01](173 kbytes)
lmoments
Version 3.04 of the L-moments code. A collection of 56 Fortran-77 routines for statistical analysis
using L-moments, and some auxiliary routines used by the L-moment routines. (submitted by J. R. M. Hosking,
hosking@watson.ibm.com) [2/Feb/90] [13/Dec/96][20/Jun/00][29/Jul/05] (279 kbytes)
lmoments.pdf
Documentation for the lmoments package (submitted by J. R. M. Hosking, hosking@watson.ibm.com)
[13/Dec/96] [20/Jun/00][29/Jul/05](217 kbytes)
lms
Models age-related reference centiles by the semi-parametric smoothing spline LMS method of Cole &
Green, Statistics in Medicine 1992;11:1305-19. Submitted by Tim Cole, (tim.cole@mrc-dunn.cam.ac.uk)
[21/Mar/96] (80k)
loess
a program for smoothing multivariate scattered data by Cleveland, Grosse, and Shyu. A pointer to how to
get loess from netlib.
logist
gkslib
lowess
see loess
lspassoc and lspcorel
Fortran77 programs for estimating the parameters and their asymtotic covariance matrix in correlation
models (lspcorel) and association models (lspassoc). Submitted by Anne-Marie MONDOT
(mondot@aurore.cict.fr). [7/Feb/00][25/Nov/02] (45k and 40k)
../S/logspline
See logspline in the S collection.
ltprsuv
The program can be applied for the simultaneous regression analysis of hazard rate and the logit of
surviving fraction with time-dependent covariates. Without the use of time-dependent covariates for the logit
regression, the model is identical to the Heckman-Walker model. Submitted by Kazuo Yamaguchi
(yama@spc.uchicago.edu) [7/Jul/98] (51k)
macanova
An Interactive Program for Statistical Analysis and Matrix Algebra. Mac and DOS executables are
available, as well as source code. Not available via e-mail. Contr. by Gary W. Oehlert (gary@stat.umn.edu)
[2/Mar/95] [mirrored nightly]
macrecur
This is an Apple Mac implementation of the regression recursive analysis diagnostics set out in Hawkins
May 1991 Technometrics. The code features interactive reordering of the cases and temporary deletions. A
binhex'ed Stuffit archive. Contr. by Doug Hawkins (doug@umnstat.stat.umn.edu) [21/May/91] (119626 bytes)
map.builder
The map() function in S permits a wide range of line and filled area maps to be drawn for use with
map-oriented data analysis. The data underlying these maps is stored in a "geographical database." The
software in this submission implements some of the ideas described in
http://research.att.com/areas/stat/doc/95.2.ps for constructing such a database. Submitted by Allan R. Wilks
(allan@research.att.com) [22/Jun/98] (71 kbytes)
markov.sas
SAS Macro code that performs competing risks analysis, using the Markov chain approach of Aalen and
Johansen. The incidence and Greenwood standard errors are computed, for any reasonable number of competing
risks. Submitted by William Anderson (William_Anderson@edwards.com) [17/Oct/00] (21 kbytes)
maxlik
A pointer to a set of Gauss programs and datasets (anotated for pedagogical purposes) to implement many
of the likelihood-based models discussed in Gary King 1989. UNIFYING POLITICAL METHODOLOGY, New York:
Cambridge University Press. [4/Jan/93]
maxsd.for
Fortran program. Tests toxicity data on dose levels d_1,...,d_k compared with a zero dose d_0 to
determine the estimated maximum safe dose. Submitted by Charles W. Dunnett
(dunnett@mcmail.cis.mcmaster.ca). [18/Apr/01] (44 kbytes)
mcem_map.sas
SAS Macro: The Maximum a posteriori Estimate using the Monte Carlo Expectation-Maximization
Algorithm. Submitted by Paul Johnson (P.Johnson@prodigy.net) [8/Dec/98][21/Jan/00][2/Jan/01] (5k)
mclust
Model-based Gaussian hierarchical clustering. (Banfield and Raftery, Tech. Rep. No. 186, Statistics
Dept., U. of Washington, Dec. 1989) Fortran source code in a shar archive. See also mclust in the S
collection. Superseded by emclust . Submitted by Chris Fraley (fraley@stat.washington.edu) [25/Sept/91]
[4/Nov/91] [19/Oct/92] [22/Jan/93][18/May/93] [17/Nov/93] [24/Mar/97] (323 kbytes)
minvol
A fortran program for the minimum volume ellipsoid estimator of multivariate location and scatter, with
robust distances. Submitted by Peter Rousseeuw (rousse@wins.uia.ac.be) [28/July/91] (30889 bytes)
mischi
A BASIC program for estimating missing cell frequencies in chi square tests for association. Submitted
by Richard Graf, Edward Alf, Jr. and Steve Williams, San Diego State University (rgraf@sunstroke.sdsu.edu)
[25/Oct/96](27k).
mk-florida
Supplemental tables from the paper, "Evaluation of a Reconstruction of the Adjusted 1990 Census for
Florida", by Michael M. Meyer and Joseph B. Kadane, in the Journal of Official Statistics. Submitted by Mike
Meyer (mikem@stat.cmu.edu) [15/Aug/96] (10k)
mpcpn
A FORTRAN program to analyze multiple-path change-point models for normally distributed data using a
Bayesian approach. Submitted by Lawrence Joseph (joseph@binky.epi.mcgill.ca), David Wolfson and Roxane du
Berger. [1/Mar/95](160k)
mpcpp
A FORTRAN program to analyze multiple-path change-point models for Poisson distributed data using a
Bayesian approach. Submitted by Lawrence Joseph (joseph@binky.epi.mcgill.ca), David Wolfson and Roxane du
Berger. [1/Mar/95](174k)
mtd
Higher order Markov chains: Estimating the MTD (Markov Transition Distribution) model [calls
NAG]. "Adrian E. Raftery" (raftery@stat.washington.edu) [11/Mar/94][25/Jul/94] (32kbytes)
MulSubSel.sas
A SAS program designed to give the subset of predictors that "best" predict all dependent variables
jointly according to the multivariate version of the Automatic Search Procedures and All-Possible-Regression
Procedures. Submitted by Ali A. Al-Subaihi (aaast11@pitt.edu). [8/Jun/00] (39k)
mvntest.gz
Multivariate Hypothesis Testing Software for Gaussian Data. C source code that compiles into a library
and several stand-alone programs. A compressed shar file. Submitted by Tapas Kanungo
(tapas@george.ee.washington.edu) [29/Feb/96] (475k)
negmul
A FORTRAN program for negative multinomial regression models for clustered event counts, plus a data
example from Sociological Methodology 1996. Submitted by Guo, Guang (gguo.ham@mhs.unc.edu) [28/Feb/96] (187k)
nlmdl
Instructions for obtaining, A C++ implementation of statistical methods in A. Ronald Galant's,
"Nonlinear Statistical Models". The source is available in several parts. (arg@ccvr1.cc.ncsu.edu)
[3/Jan/91][13/Jan/91] (3281 bytes)
nlmix
Nlmix: A Program for Maximum Likelihood Estimation of the Nonlinear Mixed Effects Model with a Smooth
Random Effects Density. This source and documentation is large. This file contains directions and more
details. Submitted by Ron Gallant, (ron_gallant@unc.edu) [17/Nov/91] [28/April/92] [7/Dec/92] [20/Dec/93]
nlmix.shar
The shar file that goes with nlmix. Please obtain the nlmix file first. (428 Kbytes)
npvar
npvar --Nonparametric estimation of semivariogram functions. Contributed by Steve Cherry
(imsgsche@bill.oscs.montana.edu) and Jeff Banfield (umsfjban@bill.oscs.montana.edu) [6/Feb/94][49 kbytes]
optim
A package of Java translations of public domain optimization routines --- fmin (1-d minimization),
Uncmin (multi-dimensional minimization), and the Minpack lmder and lmdif routines (nonlinear least
squares). A compressed tar archive. Further documentation and explanation of the routines can be found at
http://ws13.fpl.fs.fed.us/optimization.html Submitted by Steve Verrill (steve@ws10.fpl.fs.fed.us). Also
available as a self-extracting zip file here. [10/Jan/01](216 kbytes)
pair
FORTRAN program for locating cases or stimuli on a unidimensional scale given paired-comparison data (in
the form of preference ratings, point difference, or a similar interval- or ratio-level measure). Includes
user guide and sample input and output. Submitted by John Uebersax (jsuebersax@yahoo.com) [16/Dec/98] (54
kbytes)
pgam2
A SAS macro to fit the parametric conditional shared gamma frailty models, using the ML-EM algorithm
described in the Vu, Segal, Knuiman and James 2001 paper, "Asymptotic and small sample statistical properties
for shared gamma frailty models" in Communications in Statistics: Simulation and Computation, 30(3),
581-591. Submitted by Hien Thi Van Vu (hien@dph.uwa.edu.au) [9/Oct/01] (163Kbytes)
plcox
FORTRAN program to compute the parametric and nonparametric components of the relative risk function in
the partly linear Cox model. Submitted by Glenn Heller (heller@biosta.mskcc.org)[13/Oct/00] (12219 bytes)
powersas
A SAS macro to plot a power table for analysis of variance designs with any number of levels per
factor. (N410060@UNIVSCVM.CSD.SCAROLINA.EDU) [31/Oct/91] (3862 bytes)
progress
A Fortran program implementing regression methods able to identify multiple outliers. Yields least
median of squares and reweighted least squares. Includes data sets. Submitted by Peter Rousseeuw,
rousse@wins.uia.ac.be [8/June/90] (178 kbytes)
projpurs
Fortran source for Friedman's projection pursuit algorithm. [10/Oct/89] (29263 bytes)
qprob
Qprob, an algorithm for calculating the upper percentage points of the distribution of the maximum of
several studentized range statistics having independent numerators and a common denominator. For further
description, see Copenhaver and Holland, Journ. Statist. Comp. Sim., 30, 1988, 1-15. [24/July/91] Sub. by
Burt Holland (V5730E@TEMPLEVM.BITNET) (15984 bytes)
ranlib
See utexas below.
rmcmcs
A gzip'ed tar archive containing a library of routines which facilitate regenerative MCMC
simulation. Examples and documentation are included. Submitted by Anthony Brockwell
(abrock@stat.cmu.edu). [26/Mar/02] (76K)
rkpk
A shar archive of Ratfor/Fortran routines for calculating smoothing splines in regression. LaTeX
document and demo code included. (chong@pop.stat.purdue.edu) (305162 bytes) [21/May/92]
robphreg
This is an update to the ROBPHREG program that accounts for changes made in SAS Versions 8.0 and higher,
but can be used in earlier SAS versions as well. The program contains a SAS macro that uses the PHREG
procedure to fit case-cohort designs and other proportional hazards regression models with a robust
covariance matrix. When this program was first written, the robust covariance matrix could not be directly
estimated in SAS, but this feature is now available in SAS Versions 8.0 and higher. Paper reference: Barlow
WE, et al. Analysis of case-cohort designs. J Clin Epidemiol. 1999 Dec;52(12):1165-72. Submitted by William
Barlow (wbarlow@u.washington.edu), modified by ichikawa.l@ghc.org. [28/Feb/96][3/Jun/98][6/Feb/04] (62k)
rpart
C routines for recursive partitioning. Shar archive. A toolkit for new methods and a smattering of old
ones. Submitted by Terry Therneau (therneau@mayo.edu). [5/Feb/91][9/July/93][30/Aug/94][1/Dec/95](263 kbytes)
rmord
RMORD is a FORTRAN program which implements the technique of Stram, Wei, and Ware (1988, Journal of the
American Statistical Association) for regression analysis of ordered categorical repeated measurements. A
shar archive, include HP-UX executable. Submitted by Daniel Hall (dhall@pmeh.uiowa.edu) [22/Oct/96](484k)
rvbe
The package contains the data and program files in MATLAB used for the example in the paper "Modelling
Exponential Survival Data with Dependent Censoring" by Amit Bhattacharyya. Submitted by Amit Bhattacharyya-1
(Amit_Bhattacharyya-1@sbphrd.com) [13/Dec/96]
sapaclisp
collection of Common Lisp functions for the spectral analysis of time series -- See the sapaclisp
collection.
sas
The index and usage instructions for the uicstat archive of SAS code. The uicstat archive is available
via e-mail and FTP. [11/Oct/89] [11/Nov/90] (26321 bytes)
sas-mode.el
An Emacs mode for Sas (from SAS itself) [28/Aug/95] (46 kbytes)
simann
A Fortran 77 implementation of simulated annealing for the optimization of difficult and multi-optima
functions. It is useful for difficult likelihood functions and other statistical methods requiring function
optimization for estimation. See Goffe, Ferrier, and Rogers, Journal of Econometrics, vol. 60, no. 1/2,
Jan./Feb. 1994. "William L. Goffe" (bgoffe@whale.st.usm.edu), [6/Feb/94] (37 kbytes)
semml
Algorithm for the estimation of a simultaneous equation model by maximum likelihood and optimal B-robust
methods. A stacked ascii file. The procedures are presented and discussed in the paper: Krishnakumar,J. and
E. Ronchetti (1994), "Robust Estimators for Simultaneous Equations Models",to appear in the "Journal of
Econometrics". Submitted by Jaya Krishnakumar, Jaya.Krishnakumar@metri.unige.ch [1/Dec/95] (23 Kbytes)
sml
Even thoughh SML is fairly small (compiles in a few seconds) it still provides a wide range of matrix
operations including Cholesky, QR, singular value (SVD) and eigenvalue decompositions. SML is extremely
flexible. It can be configured to choose: matrix element type to be float, double, long double (or even other
types if needed); memory layout to be by-row or by-column; matrix access to be by macros or by function calls
with or without index range checking. SML offers a special storage mode that is compatible with the Template
Numerical Toolkit(http://math.nist.gov/tnt) and the book Numerical Recipes. Submitted by (qaqish AT
bios.unc.edu). [21/Oct/04] (167kbytes)
smoothwb
Smoother's Workbench for studying smoothing methods. An X windows program in Fortran and C. Lise
Manchester (lise@cs.dal.ca) [28/Jun/93] (210 kbytes)
snqn
Fast algorithms for computing two new scale estimators proposed in JASA, 88 (1993), 1273-1283, in
O(nlogn) time. File contains a Fortran function for Sn and another for Qn. Submitted by Peter Rousseeuw
(rousse@wins.uia.ac.be) and Kristof Croux (croux@wins.uia.ac.be). [5/August/94] (20 k)
snp
Instructions for obtaining, Fortran source code and users guide for a Nonparametric Time Series program
by A. Ron Gallant and George E. Tauchen. The source and user guide are separately available (about
980k). (ron_gallant@unc.edu) [3/Jan/91] [28/April/92] [20/Dec/93] [1/Mar/95] (2508 bytes)
snp.shar
Shar file to go with snp. Please obtain snp first. (480 kbytes)
snpguide.ps
Postscript documentation to go with snp. Please obtain snp first. (480 kbytes)
spgam1
A SAS macro to fit the semiparametric conditional shared gamma frailty models, using the ML-EM algorithm
described in the Vu, Segal, Knuiman and James 2001 paper, "Asymptotic and small sample statistical properties
for shared gamma frailty models" in Communications in Statistics: Simulation and Computation, 30(3),
581-591. Submitted by Hien Thi Van Vu (hien@dph.uwa.edu.au) [9/Oct/01] (157Kbytes)
spgam6
A SAS macro to fit the semiparametric shared gamma frailty models with events before study entry, using
the ML-EM algorithm described in the Vu 2003 paper, "Parametric and semiparametric conditional shared gamma
frailty models with events before study entry" in Communications in Statistics: Simulation and Computation,
32(4), 1223-1248. Submitted by Hien Thi Van Vu (hienv@unimelb.edu.au) [5/July/04] (34kbytes)
split.sas
SAS macros that implement a binary partitioning algorithm for the case of normally distributed repeated
measures response data and a continuous or ordered prognostic variable. Submitted by Mohamed Abdolell
(abdo@sickkids.ca). [31/Jul/01] (6812 bytes).
spln7
A SAS macro to fit the semiparametric conditional shared log-normal frailty models with events before
study entry, using the ML-EM algorithm described in the Vu 2004 paper, "Estimation in semiparametric
conditional shared frailty models with events before study entry" in Computational Statistics and Data
Analysis, 45(3), 621-637.Submitted by Hien Thi Van Vu (hienv@unimelb.edu.au) [12/July/04] (36kbytes)
SqcOnline
An interactive tool that is aimed at assisting practitioners with the investigation and implementation
of statistical tools in Industry. The software includes two major components: 1. Online versions of several
Military Standards that are widely used for sampling inspection (a.k.a. "acceptance sampling") and
2. Calculations of various measures that are obtained when implementing techniques that are based on runs
rules. Submitted by Galit Shmueli (gshmueli@rhsmith.umd.edu). [19/Jul/02]
ssa
A pointer to public information and data available from the U.S. Social Security Information. (2153
bytes). [12/Nov/91]
stattab
See utexas below.
symstat10.tar.gz
Procedures for SYMLOG (data input, numerical output, plots with PicTeX). Submitted by Wenzel Matiaske
(wenzel_matiaske@notes.uni-paderborn.de) [13/Aug/99](267k)
symstb
Fortran program to compute a fast numerical approximation to the Symmetric Stable distribution and
density functions. (Hu McCulloch, mcculloch.2@osu.edu) [2/Oct/95] (15k)
sweep.sas
SAS and Gauss algorithms to accompany the paper A Modified Sweep Algorithm for Interchanging between
Overparameterized and Cell Means Linear Models", by Francis Hsuan and Burt Holland. Burt Holland
(V5730E@TEMPLEVM.BITNET) [21/July/93] (22kbytes)
tmcc
The subroutine tmcc computes the upper percentage points of the extended Dunnett's
procedure. [23/Oct/91] Sub. by Burt Holland (V5730E@TEMPLEVM.BITNET) (34384 bytes)
toms717
A routine for evaluating multivariate normal integrals. From TOMS717. [17/Mar/94] Completely
unsupported, use at your own risk. There are better algorithms available from Alan Genz (4 kbytes)
toms712
A routine for generating normal deviates. From TOMS712. [17/Mar/94] Sub. by Alan Miller
(alan@dmsmelb.mel.dms.CSIRO.AU) (1 kbytes)
tpmc
To compute needed percentage points for the pairwise multiple comparison procedure in a two-way design
b1@sparc2.sta.cuhk.hk (S.H. Cheung) [1/Jun/95] (28 k)
tsa
Software and Data Sets for "Time Series Analysis and Its Applications" by R.H. Shumway & D.S. Stoffer,
Springer, 2000. Submitted by David Stoffer (stoffer@stat.pitt.edu)[10/Mar/00]
tslogit
Estimates Logan's (AJS 1996) two-sided logit (TSL) model for preferences of workers and employers,
etc. Includes manual, for DEC, Sun, Windows NT. Submitted by John Allen Logan (logan@ssc.wisc.edu)
[17/Jun/98] (692k)
ukeire
A postscript map of the United Kingdom and Eire. The map is mostly x-y pairs, so the data is useful even
if you don't have a postscript printer. (bill@stat.cmu.edu) [11/May/92] (219445 bytes).
umdhmm
This is a package for Hidden Markov Models. Currently it contains routines for forward, backward,
Viterbi and Buam-Welch estimation algorithms. Few drivers and generators allow a user to test these
routines. Submitted by Tapas Kanungo (kanungo@cfar.umd.edu) [3/Jun/98] (65 k)
utexas
Instructions and index for Fortran source code and documentation for a variety of biostatistics
applications, including group trials, randomization, and smoothing/nonparametric regression. Submitted by
Barry Brown (an12354@mdaali.cancer.utexas.edu) (2784 bytes)
validcor
This fortran program creates a 4x4 correlation matrix with valid bounds on the rest of the entries after
selecting r12, r13 and r14, insuring positive semidefiniteness. Submitted by hoffmalo@mail.armstrong.edu
[7/Jun/05] (8kbytes)
wgtreg
The program WGTREG.FOR is a FORTRAN source program for the linear regression with parametrically
weighted covariates, whose use is described by Yamaguchi (2002) in his article that appeared Sociological
Methodology 32. The manual for the use of the program is described as comments in the beginning of the source
program. Submitted by Kazuo Yamaguchi (kyamagu@midway.uchicago.edu) [3/Jan/03] (86k)
world.map
A complete world political map. Much more detailed (and much larger) than either world or
world.hires. Available only via FTP. Submitted by Becker and Wilks. [14/May/92]
xgobi
XGobi is an X Window application for interactively exploring data. Current functionality includes
brushing, identification, and the editing of connected lines, as well as rotation and the grand tour, with
several interactive projection pursuit indices. Several functions can be linked so that actions in one window
are promptly reflected in another. There is an interface to S. Its big!. Not available by e-mail, look in the
subdirectory XGobi . (dfs@research.att.com) and Dianne Cook (dcook@fisher.rutgers.edu). Mirrored daily from
http://www.research.att.com/areas/stat/xgobi/ [19/Jan/99]
xnavigation
Xnavigation is an interactive navigation program for viewing local structures in multivariate
datasets. Both manual navigation and automatic navigation which searches for smooth low-dimensional
structures are available. This is based on the author's PhD thesis. Singo Oue, (shingo@stat.cmu.edu)
[1/Mar/95] (119k)
xqz.uuen
XQz is an X Window application for interactively exploring time series data. Current functionality
includes DATA, ARMA and FREQ modules which can be linked. (mcdougal@stat.rutgers.edu) Andrew
McDougall. uuencoded, compressed, tar file. (432 kbytes) [29/Nov/94]
yamltmv.sps
An SPSS program for loglinear latent-trait Markov chain models I-A and I-B in the Yamaguchi paper that
is to appear in the 1996 Sociological Methodology. Kazuo Yamaguchi, YAMA@CICERO.SPC.UCHICAGO.EDU. [1/Dec/95]
------------------------------------------------------------------------------------------------------------------
Credit where credit is due
If you use an algorithm, dataset, or other information from StatLib, please acknowledge both StatLib and the
original contributor of the material.
------------------------------------------------------------------------------------------------------------------
Last modified: Fri Jul 29 14:00:34 EDT 2005 By Pantelis Vlachos