Some of the entries are shar archives. If you don't know how to deal with a shar archive, send the message

- submissions
- The instructions for how to submit software to statlib. [1/May/89] (1673 bytes)
- topical.index
- Topical index of the many fine software routines for S and S-PLUS which are available from StatLib. Maintained by Charles Roosen, charles@playfair.stanford.edu [28/Aug/95][28/Feb/96] (5k)
- abind
- Piece together an array from several vectors and/or arrays. This is a multidimensional generalization of the cbind and rbind functions. Written by Tony Plate (tap@comp.vuw.ac.nz) and Richard Heiberger (rmh@astro.ocis.temple.edu). [22/Oct/96] (33k)
- accflf
- Fits the very general accelerated failure model with log-F error to possibly censored data. bwb@loki.mda.uth.tmc.edu (Barry W. Brown) [2/Oct/95] (440k)
- accflf.tex
- TEX source to accompany accflf. bwb@loki.mda.uth.tmc.edu (Barry W. Brown) [2/Oct/95] (25k)
- ace.glm
- Perform ACE transformation detection on Generalized Linear Models. Extension of ace.logit. Simon Byers (byers@stat.washington.edu) [13/Dec/96] [24/Mar/97](5k)
- ace.logit
- Calculate and plot the ACE transformations for logistic regression (adapted from linear regression). Adrian E. Raftery, raftery@stat.washington.edu. [23/Jan/95] [13/Dec/96] (4k)
- adapt
- An S function that calls Genz's "adapt" suboutine for adaptive quadrature (integration) in from 2 to 20 dimensions. Adapted by Mike Meyer (mikem@stat.cmu.edu) [17/Oct/89](31709 bytes)
- addreg.tar.gz
- Additive hazards regression alternative or supplement to the Cox model. It gives plots that are shows the effect of covariates on survival. A gzipped tar file. Submitted by Harald Fekjaer (hfe@math.uio.no) [14/Oct/97] (74k) See also the Windows version
- agreg
- Obsolete, see survival2 below.
- anova
- S functions for classical and resistant analysis of averages and the analysis of variance using the sweep operator. Created by Colin Goodall (colin@jackknife.princeton.edu) [6/July/89][24/March/92](70342 bytes)
- ar.robust
- S-function to compute robust estimates of partial autocorrelation for AR processes (upto order 6). This function can also be used to obtain posterior probabilities for the competing AR models. Nhu Le (nhu@epi.terryfox.ubc.ca). [1/Mar/95] (99 k)
- ash
- A shar archive of three separate sets of programs for 1D, 2D and 3D density estimation. The last, ashn, is combined with a visualization program. David Scott, (scottdw@stat5.rice.edu) [14/Apr/93] (84kbytes)
- asypow
- Provides asymptotic sample size, power, and detectable effect calculations for a variety of generalized linear models and survival models. Shar file. bwb@loki.mda.uth.tmc.edu (Barry W. Brown) [2/Oct/95] (197k)
- asypow.dvi
- TEX dvi file for paper on asypow. Will not e-mail well. bwb@loki.mda.uth.tmc.edu (Barry W. Brown) [2/Oct/95] (47k)
- autologit
- S-plus software to perform Bayesian estimation for an autologistic model with covariates. A shar archive. Submitted by Jennifer Hoeting (jah@stat.colostate.edu) [3/Mar/99][9/Sep/99][5/Jul/00] (155k)
- avas
- A shar arhive of new S code for avas (additivity and variance stabilization). It is used to estimate transformations for regression. Contributed by Rob Tibshirani (tibs@utstat.toronto.edu) [7/Oct/89](51150bytes) {Revised, Oct 7. 89}
- balloon
- Balloon is a package of Splus functions for creating skew-t plots and winds profile plots for rawinsonde data. Submitted by Eric Gilleland (ericg@cgd.ucar.edu). [9/Nov/99] (196 kbytes)
- beran
- A shar file containing all S functions and datasets from "Statistics for Long-Memory Processes" by Jan Beran are provided along with a LaTeX file listing the data. Contributed by Brandon Whitcher (brandon@stat.washington.edu) [24/Mar/97] (124k)
- bicreg
- Bayesian model selection and accounting for model uncertainty in linear regression models, using the BIC approximation. Contributed by "Adrian E. Raftery" (raftery@stat.washington.edu) [17/Oct/94] [23/Jan/95] [13/Dec/96] [14/Oct/97] [2/Jun/98] (8 kbytes)
- bic.glm
- An S-PLUS function to implement Bayesian Model Averaging for Generalized Linear Models using BIC. Contributed by Chris T. Volinsky (volinsky@thelma.stat.washington.edu) [2/Jun/98] (19k)
- bic.logit
- Logistic regression with Bayesian model selection and accounting for model uncertainty, using the BIC approximation Contributed by "Adrian E. Raftery" (raftery@stat.washington.edu) [29/Nov/94] [28/Feb/96] [2/Jun/98] (15 kbytes)
- bic.surv
- S-Plus function to implement Bayesian Model Averaging for Cox Proportional Hazards Models, using the BIC Approximation. Includes a summary function. Contributed by Chris T. Volinsky (volinsky@thelma.stat.washington.edu) [1/Dec/95] [28/Feb/96] [13/Dec/96] [2/Jun/98] (17 kbytes)
- biv.norm
- Plots a bivariate normal density. It is a dumped file and it can be read into Splus (or R for that matter) by simply typing restore("biv.norm.dmp"). Submitted by Veith Tiemann (vtiemann@wiwi.uni-bielefeld.de) [9/Nov/99] (1053 bytes)
- bivsurv
- Functions for nonparametric bivariate survival curve analysis. Requires dynamic loading. Contributed by Ron Pruitt (ronp@umnstat.stat.umn.edu) [19/Nov/91] [27/July/92] [30/Sept/92] (112764 bytes).
- blip
- Box, Line, and Point Plot. A flexible one-dimensional distribution plot(Am.Stat. 51:353-358,1997). Shar file. lee@odin.mdacc.tmc.edu (J. Jack Lee) [2/Oct/95][3/Jun/98] (53k)
- bma
- Bayesian model averaging (BMA) to account for model uncertainty in linear regression models. Contributed by Jennifer Hoeting (jah@stat.colostate.edu) [3/Apr/95] [17/Sep/98] (27kbytes)
- boot
- Replaced by bootstrap.funs
- boott
- Replaced by bootstrap.funs
- bootstrap.funs
- S language software (bootstrap, cross-validation, jackknife), data and errata for the book "An Introduction to the Bootstrap" by B. Efron and R. Tibshirani, 1993, Chapman and Hall. Please send comments or questions to R. Tibshirani (tibs@utstat.toronto.edu) [15/June/93][18/June/93][25/Aug/93][17/Mar/94][25/Jul/94] [1/Nov/94] (116 kbytes) There is also a precompiled DOS/Windows version
- boxmat
- Boxplot Matrix. Extension of the standard oneway boxplot to cross-classified data with multiple observations per cell. A shar file which includes implementations in S and XLISPSTAT and the LaTeX paper. Richard M. Heiberger (rmh@astro.ocis.temple.edu) [19/Oct/92](138485 bytes)
- br
- A fortran program running through Splus which performs additive, robust nonparametric regression with automatic data transformation using a Bayesian Monte Carlo Markov Chain as the method of estimation. Based on Smith and Kohn, "Nonparametric regression using Bayesian variable selection", (to appear in the Journal of Econometrics.) Submitted by Mike Smith (mikes@agsm.unsw.edu.au) [28/Aug/95] [2/Oct/95] [25/Apr/96](305 Kbytes)
- brdgrun
- A shar archive of FORTRAN and Splus functions to compute shrinkage estimators with bridge penalty (\sum |\beta|^\gamma) for linear models. It uses shooting method for Lasso and modified Newton-Raphson method for general bridge. Submitted by Wenjiang FU (fuw@msu.edu) [23/May/00] (20441bytes)
- bpplot
- Box-percentile plots are similiar to boxplots, except box-percentile plots supply more information about the univariate distributions. (umsfjban@bill.oscs.montana.edu) [11/May/92] (14808 bytes)
- buffon
- The famous Monte-Carlo simulation is performed. It is a dumped file and it can be read into Splus (or R for that matter) by simply typing restore("buffon.dmp"). Submitted by Veith Tiemann (vtiemann@wiwi.uni-bielefeld.de) [9/Nov/99] (3189 bytes)
- bugs
- A list of known bugs in S, compiled by Rick Becker. (rab@research.att.com) [10/Mar/90] (32022 bytes)
- bystats
- Function to compute proportions, means, and other statistics by levels of one or more classification variables. Part of the Harrell/hmisc collection.
- calibration
- S-PLUS functions for calibration and assay development using the four parameter logistic model. The functions are particularly useful for immunoassay (RIA) and enzyme linked immunosorbent assay (ELISA) applications. moc@bdrc.bd.com (Michael A. OConnell) [17/Nov/93] (111k)
- calibration-multiple
- An addon to the calibration functions of O'Connell, Belanger and Haaland which allows reuse of the same standard curve to create numerous calibration intervals. Submitted by Jim Robison-Cox (jimrc@math.montana.edu) [28/Feb/96] (14k)
- cheatsheet
- A set of simple summaries of S syntax and expressions. Created by Barry Brown (an12354@mdaali.cancer.utexas.edu) [5/Nov/90] (8361 bytes)
- class
- Library of functions for classification. Includes robust and predictive linear and quadratic discrimination, k-nearest neighbours and allied methods. Replaces classif, and uses only C. Part of the Modapplstat collection.
- classif
- Replaced by class
- cl-s
- S routines to write out data into common lisp structures, and common lisp routines to write data in a form that S can understand. A shar archive. (rwoldford@water.waterloo.edu) [24/April/90] (22289 bytes)
- comblike
- These functions calculate the likelihood of a time series using the results of the Kalman Filter (distribution of the state given the past) and the results of the Reverse Kalman Filter (distribution of the state given the future data). Submitted by Ross Taplin (taplin@prodigal.murdoch.edu.au) [13/Jul/98] (1940 bytes)
- comprB
- A display for comparing one state (or individual) to many others. Display is described in an upcoming American Statistician. Submitted by Russell Almond (ralmond@ets.org). [1/Dec/99] (20109 bytes)
- coneplot
- Two S(plus) functions and help files to explore the use of the cone plot, a graphical object for exploring the geometric structure of high dimensional data. Brian Dawkins (brian@isor.vuw.ac.nz) [16/Jan/92] (14689 bytes).
- conf.design
- A collection of simple functions to generate and manipulate confounded and fractional factorial designs. Contr. by Bill Venables (wvenable@stats.adelaide.edu.au) [8/Aug/94] (23 k)
- convert
- translate S matrices into UNIX format sas, bmdp or blss data sets and vice versa. Also includes a function for writing an S matrix into source for xlisp. Contr. by Phil Spector (spector@gandalf.berkeley.edu) [21/May/91] (41346 bytes)
- CoxNote
- S functions for "A note on the graphical analysis of survival data" (Cox, 1979). A shar archive. Submitted by George Tomlinson (georget@utstat.toronto.edu). [22/Feb/99] (16768 bytes)
- cox_ph
- This is the shar file of GAM Cox prop hazards model, implemented in S/Splus (with most stuff in C/ratfor for efficiency). Splus driver comes with support of formula language prevalent in most other fitting software in S (lm, glm, gam etc). Rafal Kustra, rafal@utstat.utoronto. [23/Jan/95] (62k)
- cox_ph_var
- This is the shar file of Varying coefficient Cox prop hazards model, implemented in S/Splus (with most stuff in C/ratfor for efficiency). Splus driver comes with support of formula language prevalent in most other fitting software in S (lm, glm, gam etc). Rafal Kustra, rafal@utstat.utoronto. [23/Jan/95] (69k)
- cpower
- 2 Functions: cpower, Compute power of two-sample Cox or logrank survival test; ciapower, Compute power of interaction test for exponential survival; Part of the Harrell/hmisc collection.
- crr
- Software to perform subdistribution regression analysis with competing risks, using methods described in Fine and Gray (JASA, 1999). In shar format. Submitted by Bob Gray (gray@jimmy.harvard.edu). [24/May/99](44k)
- Cs
- Function to make a vector of character strings from a list of valid S names. Used most frequently for selecting subsets of variables from a data frame or from a SAS dataset. Part of the Harrell/hmisc collection.
- ctree
- Functions for construction of objects of class "ctree", which inherits from "tree". This group of functions isolates the data management capabilities of S "tree" functions. An application defining subsets of a cohort database is provided with functions to display the cohort progression plot. Contributed by Richard M. Heiberger (rmh@astro.ocis.temple.edu) [6/Oct/93]
- cut2
- An enhanced version of cut. Submitted by Frank Harrell Part of the Harrell/hmisc collection.
- CUpsy.local
- A shar file describing strategies for customizing S, with some functions and scripts used at the Columbia Univ. Psychology installation. Contributed by David H. Krantz (dhk@psych.columbia.edu) [25/Oct/90] (116284 bytes)
- d009newr
- compute the NPMLE of distribution function from doubly censored data. Contributed by Mai Zhou (mai@ms.uky.edu) [1/Jun/95] (15kbytes)
- dapply
- A dyadic version of apply() and a generalized outer product function, used to construct distance matrices, cross product matrices, etc., under general conditions. Contributed by Colin Goodall (crg@stat.psu.edu). [14/Apr/93] (23kbytes)
- dataset
- Functions that define and manipulate S "dataset" objects. A dataset is a matrix whose columns (variables) may be of different data types. Though motivated by a need to interface to SAS, they are useful in any data analysis. (therneau@mayo.edu) (Made obsolete by data.frame). [7/may/90][22/May/90] (63433 bytes)
- data.frame
- Revised and new functions and methods for handling data frames in S. These allow new classes of objects to be included as variables, improve consistency and efficiency, and make better use of classes and methods. Paper (latex) and detailed documentation included. (jmc@research.att.com) [9/Feb/94] (43k)
- data.report
- A shar archive containing the postscript version (data.ps) and the tex files to produce it for an AT&T Bell Labs reseearch report on data managment in S. (jmc@research.att.com) [30/June/92] (260430 bytes)
- date
- A shar archive of S functions for dealing with dates. The most useful of them accepts a vector of input dates in any of the forms 8/30/53, 30Aug53, 30 August 1953,..., August 30 53, or any mixture of these. Submitted by Terry Therneau (therneau@mayo.edu) [2/April/91] [12/June/91][8/July/91][23/Apr/92] [17/Aug/92][14/Apr/93] (28 kbytes)
- davison-hinkley
- A library of functions and data has been written by Angelo Canty to accompany the forthcoming Cambridge University Press publication Bootstrap Methods and Their Application by Anthony Davison and David Hinkley. Compressed shar file. (See DH/bootlib.sh.Z) Contributed by Angelo Canty (canty@masg59.epfl.ch) Mirrored periodically from EPFL Lausanne.
- ddo
- Nonlinear Calibration Design. A documented shar archive of Splus routines for simulating, displaying, evaluating, and comparing designs for nonlinear calibration experiments. Please send comments or questions to Doug Oman (oman@stat.berkeley.edu). [1/Jun/95] (215 k)
- delaunay
- S function deldir() to compute the Dirichlet (Voronoi) tesselation and Delaunay triangulation of a planar set of data points. rolf@halmos.math.unb.ca (Rolf Turner) and macq@mairi.llnl.gov (Don MacQueen) [28/Feb/96] [13/Dec/96] (87k)
- deriv3
- A modification to the Aug '91 S function deriv() including an option to generate a hessian function suitable for input to ms(). Submitted by David Smith (dsmith@stats.adelaide.edu.au) [13/Feb/92] (4694 bytes)
- Harrell/design
- Biostatistical/epidemiologic modeling, testing, estimation, validation, graphics, and prediction by storing enhanced model design attributes in the fit. Mirrored periodically from Virginia Biostat . Submitted by Frank Harrell (fharrell@virginia.edu) [31/May/94] [19/Sep/94] [2/Oct/95] [1/Dec/95][ 16/Dec/96] (939 kbytes)
- design.overview.pdf
- PDF documentation for design (see above) Submitted by Frank Harrell (fharrell@virginia.edu) [19/Sep/94] [13/Dec/96] (450 kbytes) & [1/Jun/95] (551 kbytes)
- describe
- concise printed statistical description of a vector, matrix, or data frame. Especially useful with sas.get. Part of the Harrell/hmisc collection.
- dopt
- Function to construct approximately D-optimal experimental designs from candidate points. Only simple blocking is allowed but arbitrary treatment structures are accommodated. Bill Venables (wvenable@stats.adelaide.edu.au), [1/Jun/95] (67k)
- detroff.zip
- DOS program to convert simple S help files to text format for use in S-PLUS for Windows. NOT(!) available by e-mail. Use FTP, gopher, or WWW. (you might need to "cd" to the ../DOS/S area first). Submitted by dmurdoch@mast.QueensU.CA (Duncan Murdoch). [30/Aug/94] (10k)
- discrete
- Functions d/p/q poisson and binomial, (andreas@iso.unibas.ch) [4/Nov/91] (4442 bytes)
- domouSe
- (note the capital S ...) `D'etection `O'f `M'ultivariate `O'utliers `U'sing `S' `E'nvironment. Contains 13 functions useful in detection od multivariate outliers. (ranger@iro.umontreal.ca) (Normand Ranger) [6/Oct/93] (79k)
- doshist
- This is a history function for Version 2.0 Splus for DOS. It returns a menu of past commands that can be executed or edited then executed. Jack Turnock (jturnock@afsc.noaa.gov) [30/Sept/92](3813 bytes)
- dysect
- Clustering algorithm based on dynamic altering of hierarchies. Suitable for large data sets. (See dysect in the general collection). (438 k).
- dyn.load2
- An alternative to dyn.load for S on many BSD style UNIX's. Bill Dunlap (bill@stat.washington.edu) [26/Feb/90] (28416 bytes)
- ecdf
- Empirical cumulative distribution plot. (5k) Part of the Harrell/hmisc collection. Frank Harrell (fharrell@virginia.edu) [17/Sept/93][8/Aug/94]
- ellipse
- Routines for drawing ellipses and ellipse-like confidence regions, Version 1.1 (59k) dmurdoch@pickard.mast.QueensU.CA (Duncan Murdoch) [30/Aug/94] [28/Aug/95]
- emu
- emu.v10 is an Splus object for fitting the limu model, a generalized mixed linear model for correlated responses on an arbitrary scale, with the (restricted) maximum likelihood techniques. The algorithm includes a maximization step and an updating step. The EM algorithm is used at the maximization step. References attached and comments welcome. Submitted by, Mark_X_Norleans%notes@sb.com. [1/Dec/95] (35 kbytes)
- ESS
- ESS (Emacs Speaks Statistics) is updates gnuemacs4 and is the penultimate Emacs and XEmacs user interface for programming in and controlling S-PLUS and other statistical packages (including SAS and XLispStat). A link to the primary distribution site at the University of Wisconsin.
- event.chart
- The function event.chart() is useful for plotting multiple time-to-event data at the individual level. Many extensions, such as sorting, subsetting, alignment, and multiple point and line formats, are provided. (Am.Stat. 54:60-73,2000) Submitted by J. Jack Lee (jjlee@mdanderson.org). [3/Jun/98][11/Apr/00] (73 Kbytes)
- faq
- This article contains answers to some of the most frequently asked questions in Snews. Charles Roosen charles@playfair.stanford.edu [12/Nov/92][23/Jan/95][28/Aug/95][28/Feb/96] (59k)
- faq.texinfo
- This is a TeXinfo version of the FAQ, which makes it useful as online documentation to be used with Emacs (and S/S-PLUS, as shown in the README file). Charles Roosen charles@playfair.stanford.edu [23/Mar/93][23/Jan/95][28/Aug/95] (79k)
- fda
- Superseded by mda.
- fig.old
- Device driver for interfacing to the Fig/XFig public domain interactive graphics program. Shar archive. (pd@kubism.ku.dk) [26/Feb/91] (28740 bytes) (For Splus v.2.x)
- fig
- Device driver for interfacing to the Fig/XFig public domain interactive graphics program. Shar archive. (pd@kubism.ku.dk) [24/Apr/92] [5/May/92] [24/Nov/92](17806 bytes) (For Splus v.3.x)
- find.matches
- Find matches between columns of matrices. This function is useful for drawing matched samples from a reference sample using exact matching, matching within specified tolerances, and Euclidean distance matching with specified scale constants. Part of the Harrell/hmisc collection. (11 kbytes)
- formatc
- Numeric to character conversion with flexibility of C's printf. Submitted by Martin Maechler (maechler@stat.math.ethz.ch). [19/Sep/94]
- 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 general collection of StatLib. Submitted by Chris Fraley, (fraley@stat.washington.edu) [24/Mar/91]. [12/Aug/91] [9/Oct/91] [4/Nov/91] [13/Nov/91] [24/March/92] [15/Sept/92] [22/Feb/93] [11/Mar/94] [4/May/94] (135 kbytes)
- fracpoly
- A set of functions that implement fractional polynomial models in S-Plus. Submitted by Gareth Ambler (g.ambler@ic.uk). [20/Oct/99] (29 kbytes)
- frailty.stable.ZIP
- A set of Splus function to estimate parameters of a positive stable frailty model for correlated time until event data. Described in: Recurrent Injury Event-Time Analysis. Wassell JT, Wojciechowski WC and DD Landen. Statistics in Medicine, 18(23), December, 1999, pp. 3355-3363. Submitted by James Wassell (jtw2@cdc.gov) [14/Dec/99] (9445 bytes)
- framemaker
- A shar archive containing a framemaker device driver for S. The contents are unix "diff"'s from the postscript driver in Spost.c. Submitted by David Lubinsky (david@hoqax.att.com) [22/Nov/90] (16475 bytes)
- french.intro
- Introduction to S-Plus 3.1 in French. Written for students at the Swiss Federal Institute of Technology in Lausanne. Useful as an introductory document for beginners in S-Plus. PostScript file of a LaTeX document, 51 pages. Contributed by Marcel Baumgartner (Marcel.Baumgartner@chlsnr.nestrd.ch). Contact the author for the LaTeX sources. [27/Jul/94] (988 kbytes)
- funfits
- FUNFITS is a comprehensive S-Plus module for fitting functions and nonlinear time series, including multivariate splines, Kriging and neural networks. Contributed by Doug Nychka (nychka@ucar.edu). [25/Apr/96] [24/Mar/97][24/Sep/99] (3 kbytes). The actual compressed tar file is available as funfits23.tar.gz in the S collection. Access this file via FTP, or the WWW, but not e-mail. (596k). Older version avaulable at funfits.tar.Z
- gmanova
- gmanova version 2.2 is an Splus object for simultaneous inference of multiple response variables on arbitrary scales. Maximum likelihood is used for estimating a generalized multivariate linear model and results for each response variables are combined with either Fisher's method for combining p-values or measures of the Mahalanobis distances for two different purposes. Submitted by Mark_X_Norleans%notes@sb.com. [1/Dec/95] (12 kbytes)
- gdiscr
- Nonlinear Discriminant Analysis via Scaling and ACE. A shar file. Written by Ross Ihaka (ihaka@dolphin.mit.edu) [25/May/90] (47352 bytes)
- gee
- A shar file containing S functions and a C engine for the Liang/Zeger generalized estimating equation approach to GLMs for dependent data. This revision includes support of model formulae. (carey@biostat.harvard.edu) [5/April/90][5/July/90][27/June/91] [18/July/92] [8/Aug/92] [31/May/94][22/Oct/96][3/Jun/98] (218 kbytes) There is also a precompiled DOS/Windows version
- geex
- gee.v26 is an Splus object for fitting the generalized linear regression model (GEEs) of Liang and Zeger. The syntax is similar to the glm(). This program was originally for comparing with the limu model (see emu.v10). With this program, the covariance matrix can be as large as 180x180, see Chapter 8 of Xie's dissertation for the analysis of Hummingbird learning data. Submitted by Mark_X_Norleans%notes@sb.com. [1/Dec/95] (32 kbytes)
- gendat
- This S function implements the algorithm of Thompson and Taylor for the generation of pseudo data points drawn from the estimated density of an observed data set. Created by Barry Brown (an12354@mdaali.cancer.utexas.edu) [30/Aug/89](8773 bytes)
- gibbsit
- Determines the number of iterations required to achieve a specified accuracy level in Markov chain Monte Carlo. An S translation of gibbsit in then general collection. Submitted by Steven Lewis (slewis@stat.washington.edu) [1/Mar/95] [4/Oct/95](7 kbytes)
- glm
- A shar archive of glm code for new S. The core is a Newton minimizer. The modular structure allows easy user extension to "nonstandard" likelihoods. (A different version of glim). (chong@popo.stat.purdue.edu) [6/Feb/91] (42753 bytes)
- glib
- Bayesian generalized linear modeling, model comparison and accounting for model uncertainty. An S-PLUS function. Adrian Raftery, (raftery@stat.washington.edu) [17/May/93] [18/May/93] [13/Dec/96] [2/Jun/98] (33 Kbytes)
- glim
- Fit one of the models described in McCullagh and Nelder, "Generalized Linear Models". Contributed by Terry Therneau (therneau@mayo.edu) [27/Sept/89][Updated 20/June/90,19/Nov/90,29/Jan/91] (22505 bytes)
- glimdesign
- Functions to allow design matrices to be specified in S using Glim or Genstat-type structure formulae. For example, design(a*b+x,length(a)), generates a (suitably labelled) design matrix for a double classification with covariate x. (jreimann@stat.berkeley.edu)[31/Mar/90] [16/Jan/92] (20370 bytes)
- gnuemacs4
- See ESS above.
- gnutags
- A version of the gnuemacs tags program which has been modified to find S function definitions. You can use this to find function defintions or arbitrary strings in multiple related files. Submitted by David Lubinsky (hoqas!hoqas2!david@hoqax.att.com) [4/Sept/90] (41607 bytes).
- graff
- Graphical analysis of fractional factorial designs. A shar archive. Contributed by Perry Haaland (pdh@bdrc.bd.com) and Michael O'Connell (moc@bdrc.bd.com). [10/Oct/90] (70041 bytes)
- hare
- Hazard Regression: S functions to estimate the conditional log-hazard function based on possibly censored data and covariates and obtain corresponding densities, hazard rates, probabilities, quantiles and random samples. Includes proportional hazard model as special case. Charles Kooperberg (clk@fhcrc.org) shar archive. [26/April/93][17/Aug/93][1/Dec/95] [28/Feb/96] [24/Mar/97] [3/Jun/98] (193 kbytes)
- haerdle
- S and C code from `Smoothing Techniques with Implementation in S' by Wolfgang Haerdle, Springer, 1991. A shar archive. Contributed by Brian Ripley (ripley@stats.ox.ac.uk) [30/March/91][7/April/92] [8/Aug/92][21/Feb/94] (124 kbytes) There is also a precompiled DOS/Windows version
- Harrell/hmisc
- The Hmisc library contains many functions useful for data analysis, high-level graphics, utility operations, functions for computing sample size and power, translating SAS datasets into S-PLUS, imputing missing values, advanced table making, variable clustering, and character string manipulation. A shar archive. Mirrored periodically from Virginia Biostat . Contributed by Frank E Harrell Jr (fharrell@virginia.edu) [13/Dec/96] (567k) See also the DOS version
- hbdp
- Modifications of the high breakdown point estimators LMS, LTS, and MM for logistic regression, Poisson regression or Gamma regression with large groups. Submitted by Andreas Christmann (a.christmann@hrz.uni-dortmund.de) [10/Sep/1998] (7k)
- hdr
- A shar archive of functions for estimating highest density regions in one and two dimensions. Contributed by Rob Hyndman (hyndman@stats.mu.oz.au).[29/Nov/94] [3/Apr/95](16 kbytes)
- heft
- Hazard Estimation with Flexible Tails: S functions to estimate an unknown hazard function based on possibly censored data and obtain corresponding densities probabilities, quantiles and random samples. Charles Kooperberg (clk@fhcrc.org)shar archive. [26/April/93][17/Aug/93][9/Oct/93] [31/May/94] [1/Dec/95] [24/Mar/97] [3/Jun/98](141 kbytes)
- hessian
- Finite Difference Hessian Approximation from Function Values. An S implementatoin of Dennis and Schnabel's FDHESSF routine. Submitted by Barry Brown (an12354@mdaali.cancer.utexas.edu). [30/May/90] (26302 bytes)
- hmm
- A shar file containing a S library to fit continuous-time hidden Markov models to irregularly spaced longitudinal observations. Contributed by Alexandre Bureau (abureau@stat.berkeley.edu) [24/May/00] (271k)
- hoeffd
- Function to compute Hoeffding's D statistic for testing independence of two variables, along with asymptotic P-values. Does computations for a matrix of variables, using pairwise deletion of NAs. Part of the Harrell/hmisc collection.
- hyperstrat
- Functions for determining exact p-values and confidence bounds from stratified samples using the hypergeometric distribution. "John P. Wendell" (WENDELL@busadm.cba.hawaii.edu) [2/Oct/95] (18k)
- impute
- Class and methods for imputation of missing values Works with the describe function (counts no. of imputed values) and the transcan function (develops imputation models predicting each predictor variable from all other predictors). Part of the Harrell/hmisc collection.
- imsl
- A shar archive containing New S code for using IMSL routines from S. Includes a patch for error handling and print routines, a Makefile, and examples of interface routines and S functions. Created by Ritei Shibata (shibata@statlab.math.keio.ac.jp) [6/May/89](21202 bytes)
- hazcov
- A shar archive giving S functions for hazard smoothing with multiple covariates. Contributed by Robert Gray (gray@jimmy.harvard.edu). [16/Oct/97] (44k)
- imsl2
- Simpler IMSL interface to new S than the previous "imsl". The new dynamic loading routine dyn.load2, contributed by Bill Dunlap, is necessary. Necessary subroutines will be directly searched for the IMSL library. A shar archive. ( shibata@math.keio.ac.jp ) [14/Dec/89] (19718 bytes)
- integrate
- Numerically integrate a function of a single variable. Includes S functions and Fortran and C source code. Created by Barry Brown (an12354@mdaali.cancer.utexas.edu) (Made obsolete by integrate2). [Updated 6/Nov/90] (81709 bytes)
- integrate2
- Single and multiple dimension numeric rectangular integration. (Change: Add good multiple dimension routine; improve packaging.) Includes S functions and Fortran and C source code. Shar file. bwb@loki.mda.uth.tmc.edu (Barry W. Brown) [2/Oct/95] (190k)
- intersect
- S and Splus routines to compute projections, intersections, orthogonal complements, gram.schmidt extensions, etc. for glm and lm model matrices. Brian Junker (brian@stat.cmu.edu), [23/Jan/95] (22k)
- interval.tar.gz
- Compressed tar file with 3 functions for use with interval censored data, to fit nonparametric survival curves, plot them, and do logrank or Wilcoxon type tests. See also windows version. Submitted by Michael P. Fay (m7f@helix.nih.go). [6/Apr/99] (11k)
- invmf
- (Invert Monotone Function). Finds x such that f(x,...)=y for f montone in x and specified y. shar file bwb@loki.mda.uth.tmc.edu (Barry W. Brown) [2/Oct/95] (37k)
- itsim
- Routines for output analysis of multiple sequences of iterative simulation (e.g., the Gibbs sampler or the Metropolis algorithm) (gelman@gandalf.berkeley.edu) [3/May/92][28/Feb/96] (9 kbytes)
- invrg
- an implementation of Duan and Li's (1991) inverse regression, handling multiple slices. Written entirely in S. Duan and Li, Slicing regression: A link free regression method. Ann. of Statist. (1991) 505-530. Mai Zhou (mai@ms.uky.edu) [4/May/94][28/Aug/95] (2685 bytes)
- isomds
- Functions for Kruskal's isotonic multidimensional scaling and Shepard plots. Part of the Modapplstat collection.
- jpn
- A shar archive containing S functions to plot the coast-line of Japan and prefecture boundaries; similar to the usa() function. Contributed by Ritei Shibata (shibata@statlab.math.keio.ac.jp) [6/May/89](73845 bytes)
- kalman
- A pair of S functions, named filter and forecast, implementing Kalman filter recursions using the EM algorithm. nantig@fi.gs.com (Gary Nan-Tie) [17/Nov/93] (9k)
- kappa
- The function kappa.cfint() will compute four different confidence intervals for the interrater agreement kappa in a 2x2 table. jjlee@odin.mda.uth.tmc.edu (J. Jack Lee), [17/Oct/94] (11k)
- kde
- Two-dimensional kernel density estimation with help page. Submitted by G.P.Nason@bristol.ac.uk. [18/June/93] (10kbytes)
- kde2d
- Two-dimensional gaussian kernel density estimation. Features: scaling, spherizing, automatic bandwidth, 1st derivatives. Submitted by G.P.Nason@bristol.ac.uk, and Martin Maechler (maechler@stat.math.ethz.ch)[29/Nov/94] (16 kbytes)
- kernel
- S-PLUS/Fortran functions for implementing several of the procedures described in the book `Kernel Smoothing' by M.P. Wand and M.C. Jones, London: Chapman and Hall, 1995. Contributed by Matt Wand (wand@agsm.unsw.edu.au). [28/Feb/96] [25/Jan/99] (139k)
- koq
- Calculates Kent & O'Quigley's measure of dependence for censored data, when dependence between time and covariates is modelled by Cox's model. Function definion and help file. Contributed by Andrej Blejec (andrej.blejec@uni-lj.si) and Janez Stare (janez.stare@mf.uni-lj.si) [13/Dec/96] (13k)
- lasso
- Functions for estimation of linear, generalized linear and the proportional hazards models via the "lasso" selection method. submitted by Rob Tibshirani (tibs@utstat.toronto.edu) [23/Jan/95] (501 kbytes)
- lasso2
- A shar archive of C and S-PLUS functions to solve regression problems in which the L1 norm of the estimated parameters is constrained. Currently handles normal linear models and generalised linear models. It uses S-PLUS modelling functions. Submitted by Berwin Turlach (bturlach@stats.adelaide.edu.au). [4/Nov/98][14/Dec/99] (954 kbytes)
- latex.table
- Convert an S matrix to a LaTeX table, with major groupings, decimal alignment, multi-page tables, and control of number of digits to right of decimal point. See also `mat2tex'. Supersceded by print.display. Frank Harrell (fharrell@virginia.edu) [6/Oct/92][17/Sept/93] (16kbytes)
- linkinf
- Two S-PLUS functions to compute influence diagnostics for assessing case influence on the goodness-of-link test in logistic regression. Submitted by Andy H. Lee (a_lee@bligh.ntu.edu.au) [28/Feb/96] (4k)
- linreg
- LINREG contains 9 functions useful in linear regression for diagnostics, residual plotting, Box and Cox transformations, an Anova table, bootstrap of a linear regression and a sequential ANOVA table. A shar archive. Submitted by Christian Leger (leger@iro.umontreal.ca) [14/Sept/90] [23/Oct/91] (31897 bytes)
- lmix
- replaced by nlme
- lmwsltws
- Modifications of the high breakdown point estimators LMS and LTS proposed by P.J. Rousseeuw for logistic regression and Poisson regression with large groups. Written by Andreas Christmann (a.christmann@hrz.uni-dortmund.de) [2/Jun/98] (11k)
- logist
- Superseded by design
- logspline
- S functions to, fully automaticly, estimate an unknown density based on possibly grouped or censored data and obtain corresponding probabilities, quantiles and random samples. C Kooperberg (clk@stat.washington.edu) [5/Feb/92][6/Oct/92][26/April/93][17/Aug/93][6/Oct/93] [1/Dec/95] [24/Mar/97] (182 kbytes)
- lsdensity
- Replaced by logspline (above)
- lspline
- Functions for L-spline smoothing using a penalty term defined by an arbitrary linear differential operator L. Contributed by Jim Ramsay James Ramsay (ramsay@cict.fr) [14/Oct/97] (640k)
- lsother
- For both Windows and Unix ls() gives unquoted output that is void of any names beginning with a period. ls(<simplesubstring>) (optional quotes) works the same in Windows and Unix, giving all object names that contain the simple substring. John R. Wallace jw@u.washington.edu, [3/Apr/95] (3kbytes)
- lspec
- Logspline estimation of a possibly mixed spectral distribution: S functions to estimate the spectral density, line spectrum and spectral distribution based on time series data. C Kooperberg (clk@stat.washington.edu) [26/July/93][17/Aug/93][9/Oct/93][1/Dec/95] [24/Mar/97][3/Jun/98] (95k kbytes)
- ../general/map.builder
- See mar.builder in the General Archive
- maps.fips
- Functions and a database, to be used with the map library, for maps based on additional search keys, such as fips codes and state abbreviations. Contributed by Colin Goodall (crg@stat.psu.edu). [14/Apr/93] (147 kbytes)
- matrogram
- S-Plus program for a two-dimensional representation of a dendrogram. Submitted by Marcel Hendrickx (hendrickx.marcel@compaqnet.be). [21/Jan/00] (12kbytes)
- mclust
- Model-based cluster and discriminant analysis, including hierarchical clustering and EM for parameterized Gaussian mixture models. Fortran code interfaced to Splus (UNIX and Windows). Mirrored daily from ftp.u.washington.edu. Submitted by Chris Fraley (fraley@stat.washington.edu) [25/Sept/91] [21/Oct/91] [4/Nov/91] [16/Jan/92] [13/Feb/92] [19/Oct/92] [22/Jan/93] [17/Nov/93] [24/Mar/97] [26/Aug/98]
- mclustem.one
- Classification of binary Gaussian + Poisson mixtures in the plane. Fortran with Splus interface. Submitted by Chris Fraley (fraley@stat.washington.edu). [24/Mar/97] [14/Oct/97] (17k)
- mcmcirt
- Markov chain Monte Carlo code for estimating item response theory (IRT) models, including 2- and 3-parameter logistic models, generalized partial credit models, and models with rater effects. A shar archive. Submitted by Richard J. Patz and Brian W. Junker. Please send comments or questions to R. Patz (rpatz@ctb.com). [8/Sep/99] (149k). See also the windows version.
- mda
- Mixture Discriminant Analysis. Tools for fitting mixture models for classification, with subspace selection as in LDA. This archive supersedes fda, and includes revisions of all the functions there. Submitted by Trevor Hastie and Rob Tibshirani (trevor@playfair.stanford.edu) [22/Mar/96] (151k)
- menusys
- Menu system with generic menubar and popup menus; has methods for Emacs and can be extended to other systems (e.g., Tcl/Tk). A shar file. Submitted by Jeff Hallman (m1jjh00@FRB.GOV). [3/Jun/98] (36 Kbytes)
- merge.data.frame
- SPlus function to merge corresponding rows of two data frames sorted over the values of the set of like-named columns. Contributed by Raymond W. FALK (ray@fred.med.yale.edu) [26/Oct/94] (11k)
- minfun
- A modification for S of David Gay's routine, mnfb, that finds parameter values to minimize a user supplied function. Created by Barry Brown (an12354@mdaali.cancer.utexas.edu) [10/Oct/89](171388 bytes)
- mixreg
- A package of Splus functions to compute the fit of a mixture of linear regressions and implement various associated inferential procedures. Submitted by Rolf Turner (rolf@math.unb.ca). [15/Sep/98][2/Oct/98][20/Oct/98] (55k)
- mmreg
- Robust regression estimator (MM-estimator). ("Plain Splus" without any extra C / Fortran code.) Andreas Ruckstuhl ruckstuhl@stat.math.ethz.ch [28/Aug/95] (37 Kb)
- Modapplstat
- Software, scripts, and data from the book Modern Applied Statistics with S-Plus by W.N. Venables & B.D. Ripley. There are unix archives and a Windows version. This material is not available via e-mail. Mirrored from Brian Ripley's collection at Oxford. (ripley@stats.ox.ac.uk). Previous version is also available here.
- modelsum
- A tabular summary of functions introduced in "Statistical Models in S". LaTeX document, 6 pages. (zaslavsk@hustat.harvard.edu) [18/March/92] (10151 bytes)
- model-plot
- Set up plots for testing model predictions vs. data. Frank E. Ritter (ritter@psychology.nottingham.AC.UK) [4/Jan/1994] (17k)
- monsmo
- Performs a monotone smoothing on the vector x, optionally using weights in a parallel vector wt. Created by Barry Brown (an12354@mdaali.cancer.utexas.edu) [1/Sept/89](15812 bytes)
- motif.print.sh
- Prints contents of motif() or openlook() window to colour postscript file. Contributed by Thomas Lumley (thomas@biostat.washington.edu) [14/Oct/97] (7k)
- msrob
- A shar file of S-plus functions for robust regression model selection using stochastic complexity (minimum description length) criterion, robust AIC and robust BIC. Submitted by Guoqi Qian (qian@neptune.stats.latrobe.edu.au) [31/Aug/98](33349 bytes).
- multiv
- Library of functions (S functions, Fortran source code) for hierarchical clustering (includes O(n^2) implementations, and 'see-it- as-it-happens' visualization program), partitioning, bond enery algorithm, Sammon mapping, PCA and correspondence analysis; latter includes fuzzy coding function. Contributed by Fionn Murtagh (fmurtagh@eso.org) [29/Aug/94](205 kbytes)
- multinom
- A shar archive of functions to provide a model class for fitting log-linear multinomial models without using surrogate Poisson models. Also includes an enhanced nnet library (see Modapplstat). Uses C. Part of the Modapplstat collection.
- negbin
- Functions for fitting and analysing Negative Binomial GLMs. Bill Venables, venables@stats.adelaide.edu.au (33k) [30/Aug/94]
- nesi
- New Environment for Statistical Inference. Various S functions for Classical Inference combined with graphical presentations of the results. A tree structured menu allows the user to interactively selection the method. This is an enhanced version (version 3.0) of the old "nesi". (shibata@math.keio.ac.jp) [1/Nov/89][10/Mar/90](177k bytes)
- nlrchoice.tar
- Splus-Tools for Model Selection and Construction of Bootstrap Confidence, Prediction and Calibration Intervals in Heteroscedastic Nonlinear Regression A tar file (so only available via FTP and WWW) Submitted by Joerg Polzehl (polzehl@wias-berlin.de) [24/Mar/97] (266k)
- nls2
- A set of SPlus functions and programs to estimate the parameters of a non-linear regression model over a given set of observations. This file contains instructions for retrieving the code by ftp, gopher or WWW. Bouvier (ab@jouy.inra.fr) [29/Mar/94]
- nlme
- Functions, classes, and methods to fit and analyze linear and nonlinear mixed-effects models. A gzip'ed tar archive. Contributed by Jose' Pinheiro (jcp@research.bell-labs.com) and Douglas Bates (bates@stat.wisc.edu).[3/Jun/93] [6/Feb/94] [1/Mar/95] [2/Aug/96] [16/Oct/97] [25/Jun/99][2/Aug/99](1.8Mbytes) There is also a precompiled DOS/Windows version
- nnclean
- A function to find features in point processes based on nearest neigbour distances. Simon Byers, (byers@stat.washington.edu) [5/Jun/96][3/Jun/98] (5k)
- nnet
- Model class for feed-forward neural networks with fitting and prediction functions. Superseded by Modapplstat.
- nonlin
- Nonlinear regression code. A corrected version of material from Appendix 3.1.1 of Bates and Watts, "Nonlinear Regression Analysis and Its Applications". A shar archive, Doug Bates (bates@stat.wisc.edu) Fixes by Peter Dalgaard (pb@kubism.ku.dk) [25/Oct/90] [22/Nov/90] (5054 bytes)
- ogrep
- Function to do text searches in data and function objects in S. Versions for Unix and MS Windows. Contributed by Duncan Murdoch,dmurdoch@pickard.mast.QueensU.CA [23/Jan/95] (10 kbytes)
- oswald
- Oswald is an S-plus library for the analysis of Longitudinal Data. It includes informative dropout modelling and GEE analysis. A compressed tar archive. Contributed by David M Smith (D.M.Smith@lancs.ac.uk) [13/Dec/96] (390k)
- oz
- A shar archive containing S functions to plot the coast-line of Australia and state boundaries; similar to the usa() and jpn() functions. Contributed by Bill Venables (wvenable@spam.ua.oz.au) [4/Aug/92] [1/Jun/95](66 kbytes)
- pathogen
- This file contains the source codes for the S-Plus functions described in Levesque, A.C. and Eaves, D.M., Decision Modelling for Lot Acceptance, Chapter 9 in Advances in Plant Pathology (Springer-Verlag). pp 241-277. Submitted by David Eaves (eaves@cs.sfu.ca) [8/Oct/98][29/Oct/98][30/Oct/98] (15k)
- pav
- An implementation of the one-dimensional pool adjacent violators algorithm for monotone regression. Submitted by Anthony Gamst (acgamst@osiris.ucsd.edu). [10/May/00] (10k)
- pear
- Splus 3.2 functions which implement the methodology for periodic autoregression modelling discussed in my article: A. Ian McLeod (1994) "Diagnostic Checking of Periodic Autoregression Models with Application", Journal of Time Series Analysis pp.221--233. Queries to aim@uwo.ca. [3/Apr/95](57kbytes)
- ploteff.tar
- Splus methods for visualizing non-linear functions (e.g., neural networks, Gaussian process models, and bagged trees) through plots of "additive effects". Tar format. Submitted by Tony Plate (tplate@ibm.net) [4/Feb/99] (71kbytes)
- plr
- Polytomous logistic regression: fit multinomial response models with possibly cross-classified outcomes, also describable as conditional loglinear models. Submitted by Alan Zaslavsky (zaslavsk@hcp.med.harvard.edu)[3/Jun/98][31/Jan/00] (215 kbytes)
- pls
- Fortran, Matlab and Splus code for Univariate Partial Least Squares regression as described in Denham(1994), "Implementing partial least squares", (to appear in Statistics and Computing). Submitted by Mike Denham (snsdenhm@reading.ac.uk) [85 kbytes]
- pmw
- printer-menu-window, a shar archive. A revised printer driver for new S. Emulates graphics windows and pop-up menus under Emacs on an ordinary 24 line ASCII terminal. Contributed by Richard M. Heiberger (rmh@gaia.stat.temple.edu) [25/May/89](64813 bytes)
- polyclass
- A set of S functions for classification and polychotomous regression using adaptively selected polynomial splines for the model building. Charles Kooperberg (clk@fhcrc.org) [24/Mar/97][3/Jun/98] (170k)
- polymars
- A set of S functions for classification and (polychotomous) regression using adaptively selected polynomial splines for the model building. Contributed by Charles Kooperberg & Martin O'Connor, (martin@stat.washington.edu). [15/Oct/97] (372k)
- polynom
- a small collection of S functions to implement a class for univariate polynomial manipulations. Bill Venables (wvenable@stats.adelaide.edu.au) [28/Feb/96] (50k) There is also a precompiled DOS/Windows version
- postscriptfonts
- Functions to plot strings under postscript() with mixed fonts, character sizes, sub- and super-scripts; and display of available postscript fonts. Updated for Splus 3.3. (zaslavsk@hustat.harvard.edu) [27/April/92] [7/April/93] [17/Sept/93] [28/Aug/95] [13/Dec/96] (48 kbytes)
- power
- A shar archive containing S functions for power and sample size analysis for tests of the general linear hypothesis. Contributed by Daniel F. Heitjan (dheitjan@biostats.hmc.psu.edu). [15/May/91][27/Nov/91] (15501 bytes).
- pr
- A shar archive containing S functions which will print any object as a list. Simple but useful for checking the organization of an object. Contributed by Ritei Shibata (shibata@statlab.math.keio.ac.jp) [6/May/89][Updated 23/Aug/91] (1449 bytes)
- principal.curve
- A shar archive containing S functions which fit a principal curve a matrix of points in arbitrary dimension Trevor Hastie (trevor@stat.stanford.edu) [Aug/7/93] (15k)
- primm
- A shar archive containing S functions which fit a hierarchical Poisson regression model. Poisson Regression Interactive Multilevel Modeling (PRIMM) Cindy L. Christiansen and Carl N. Morris Cindy Christiansen (cindy@hustat.harvard.edu) [22/Oct/96] (38k)
- print.display
- A set of object-oriented S functions that harness the automatic printing facility in S to convert an S object to another format, e.g. displaying an S object in its own window, converting a data.frame to a LaTeX table, for displaying an S help file in a LaTeX window. Richard M. Heiberger (rmh@astro.ocis.temple.edu) and Frank E. Harrell, Jr. (fharrell@virginia.edu). [4/May/94] (130 kbytes)
- psglm.tar.gz
- S functions for doing projected scores (approx conditional inference) for GLM's with independent strata. A compressed tar archive. Contributed by Richard Waterman (waterman@compstat.wharton.upenn.edu) [24/Mar/97] (13k) See also the Windows version
- pspline
- A shar archive containing S functions which implement penalized spline smoothing. Contributed by Jim Ramsay (ramsay@ramsay2.psych.mcgill.ca) [2/Aug/96] [13/Dec/96] (78k)
- quadratic
- S functions for quadratic programming: see Goodall and Brown (1994), In: A Probability, Statistics and Optimization, A Tribute to Peter Whittle (Wiley). Contributed by Colin Goodall (crg@stat.psu.edu). [1/Mar/95] (8 k)
- quadprog
- A shar archive containing FORTRAN77 and S-PLUS functions for solving quadratic programming problems. Contributed by Berwin Turlach (bturlach@stats.adelaide.edu.au) [14/Oct/97][6/Jul/98][23/Jul/98] (117k)
- quantreg
- Routines to compute quantile regressions including analogues of trimmed means for the linear model. Updated version contains new functions implementing new forms of rank tests for linear models based on the dual quantile regression process. (roger@ysidro.econ.uiuc.edu)[23/Aug/91] [1/Jun/95](48 kbytes)
- rational
- A few small functions to find numerical rational approximations using an obvious continued fraction method. Created by Bill Venables (wvenable@spam.ua.oz.au) [7/Aug/89](12607 bytes)
- rcorr
- function to compute a Pearson or Spearman correlation matrix based on pairwise deletion of missing data Part of the Harrell/hmisc collection.
- recipes
- A provisional and largely untested interface between S and routines from "Numerical Recipes in C". The interface routine include methods for root-finding, optimization and so forth. A shar archive. (brian@toru.isor.vuw.ac.nz) [22/April/90] (90230 bytes)
- reldist
- A program that implements relative distribution methods (comparison density). Includes non-parametric estimates of the relative density, CDF and relative polarization indices. Submitted by Mark S. Handcock (handcock@stat.psu.edu) [23/Oct/98][7/Sep/99](47k)
- repeated
- Ian Abramson's S-program and manual for likelihood-free repeated measures/growth curve fits; handles imbalance, multiple fixed/floating effects, dependencies. (abramson@osiris.UCSD.EDU) [21/Dec/89] (30231 bytes)
- rg-corr
- This is a set of four fumctions in SHAR file that computes the (robust) Greatest Deviation Correlation Coefficient that is usefulon normal and erratic data. (ma_rg@selway.umt.edu) [31/Oct/91] (4032 bytes)
- rg-regress
- A Robust estimate of the intercept, slopes, and error scale in a simple or multiple linear regression with the Greatest Deviation Correlation Coefficient. Written by Rudy Gideon (ma_rg@selway.umt.edu) [5/May/92] (4813 bytes)
- ripley
- Functions and datasets used for teaching at Oxford. Used in sguide. A shar archive. Submitted by Brian Ripley (ripley@stats.ox.ac.uk) [7/Aug/92] [17/May/93][21/Feb/94] [19/Oct/94] (71 kbytes)
- rm.tools2
- S-plus functions for the graphical, parametric, non-parametric and likelihood analysis of repeated measures (i.e. longitudinal data). Contributed by Adelchi Azzalini (adelchi@stat.unipd.it) and Monica Chiogna (monica@stat.unipd.it). [26/Jan/94][28/Feb/96] (93k) See also the Windows version
- rm.tools2.ps
- Postcript source of user manual for rm.tools. (552k).[26/Jan/94][28/Feb/96]
- rms.curvatures
- Root mean square relative curvature measures for nonlinear regression. Bill Venables (wvenable@stats.adelaide.edu.au), [9/July/93] [31/May/94] (6 kbytes)
- rmult
- Produces a matrix in which each row is an independent drawing from a specified multinomial distribution. Submitted by Jonathan Rougier (J.C.Rougier@durham.ac.uk)[3/Jun/98](6 kbytes)
- robeth
- Fortran code and S interface for the ROBETH family of robust statistics functions documented in `Algorithms, Routines and S Functions for Robust Statistics' (Wadsworth & Brooks/Cole, 1993) by A. Marazzi. This file contains instructions for retrieving the code by ftp, or WWW. Contributed by Alfio Marazzi (Alfio.Marazzi@iumsp.unil.ch) [16/Mar/94] [24/Mar/97]
- robust
- Collection of S functions for robust statistics plus a model class for robust linear regression via Huber's M-estimator with valid standard errors for estimates. Also contains some `dirty' datasets. Superseded by Modapplstat.
- rowmeans
- Row and column means, sums, variances, calls C for speed. Submitted by Tim Hesterberg (T_Hesterberg@FandM.edu). [19/Sep/94] (10k)
- rpart
- A shar archive of functions for recursive partitioning - an alternative to tree(). Terry Therneau & Beth Atkinson (atkinson@mayo.edu) [14/Oct/97] [3/Jun/98] (1.2M)
- rpart2
- A shar archive of functions for recursive partitioning - an alternative to tree(). Update of rpart1, including case weights. Runs on Splus5.1. Terry Therneau & Beth Atkinson (atkinson@mayo.edu) [8/Jun/00] (1.3M)
- rrr
- Splus function for reduced-rank regression and softly shrunk reduced-rank regression. Submitted by Magne Aldrin (magne.aldrin@nr.no). [19/Apr/99][8/Mar/00] (14k)
- s.f77.c.libs
- Simple interface from S to f77 or C programs which use Fortran I/O. No modifications to either S or IMSL are required! IMSL and some other libraries use Fortran I/O for errors. Partial replacement for dyn.load() when dyn.load is unavailable. (rmh@gaia.stat.temple.edu). [27/Nov/89] (7775 bytes)
- safe.predict
- Modified versions of bs() and ns() that allow safe predictions, especially in the context of the S modelling functions. New predict() methods as well. Trevor Hastie (trevor@research.att.com) [11/Mar/94] (14k)
- samplesize-betabin
- Calculates optimal sample sizes for a beta-binomial mixture model in a situation where both priors and utilities differ for two parties in a negotiation process, Based on the paper Wolfson, L.J., Kadane, J.B., and Small, M.J. (1995), "Expected Utility as a Policy-Making Tool: An Environmental Health Example", Lara Wolfson (lara@stat.cmu.edu) [1/Jun/95]
- salsa
- A shar file containing S-plus functions, C and fortran codes,which provide curvilinear principal components analysis (a particular nonlinear PCA) via a Smoothing Alternating with Least Squares Algorithm (SALSA). For more details, see Besse and Ferraty, 1995, Ferraty, 1997. Contributed by Frederic Ferraty, (ferraty@cict.fr) [14/Oct/97] [3/Jun/98] (170kbytes)
- samplesize-mean
- An S-plus program to calculate sample sizes based on highest posterior density intervals for normal means using three different Bayesian approaches. Submitted by Lawrence Joseph (joseph@binky.epi.mcgill.ca) and Patrick Belisle. [7/Sep/95] (83 kbytes)
- samplesize-prop
- A S-plus program to calculate sample sizes based on highest posterior density intervals when comparing two binomial proportions using three different Bayesian approaches. Submitted by Lawrence Joseph (joseph@binky.epi.mcgill.ca), Roxane du Berger and Patrick Belisle. [7/Sep/95](53kbytes)
- sas.get
- an update to the S-Plus function for converting SAS datasets to S data frames or lists. Part of the Harrell/hmisc collection.
- sas.vs.S
- A short (and somewhat outdated) comparision of Sas and S. (therneau@mayo.edu) Terry Therneau, [20/Dec/93] (11 kbytes)
- sbayes
- S functions for approximate Bayesian analyses. (luke@umnstat.stat.umn.edu)[12/Aug/90] [21/Jan/92] Updated by Alix Gitelman (alix@stat.cmu.edu) [28/Feb/96] (191k)
- scat1d
- One-dimension scattergram (generalized rug()). Part of the Harrell/hmisc collection.
- screen
- A shar archive containting functions to make use of X or Suntools windowing when editing or getting help. The archive contains substitute versions of help and ed, and new functions tool.options and textedit. Contr. by (zaslavsk@dolphin.mit.edu) [24/May/89](6453 bytes)
- sguide.ps1 sguide.ps2
- `Introductory Guide to S-Plus'. A beginners' guide to doing statistics in S-Plus. SGuide.ps1 and SGuide.ps2 are PostScript fullsize and reduced 2-on-1 respectively. Archive ripley is also needed. Submitted by Brian Ripley (ripley@stats.ox.ac.uk) [7/Aug/92] [19/Oct/92] [21/Feb/94][19/Oct/94] (sguide.ps1 485 kbytes; .ps2 474kb)
- simplex
- The Nelder-Mead simplex algorithm for function minimization. The method uses no derivatives, is very robust in the face of local minima, but is slow. A shar file. Submitted by Terry Therneau (therneau.terry@mayo.edu). [15/Nov/99] (14 kbytes)
- sir
- S-Plus function for Sliced Inverse Regression based on the algorithm described by K.C. Li (1991, JASA 86, 316-327). Either the number of slices or the elements per slice can be controlled. Contributed by thomas@wiwi.hu-berlin.de (Thomas Koetter) [28/Aug/95] (4710 bytes)
- slab
- A lab course for teaching probability and statistics includes a postscript lab manual and some new functions which make S easier for students to use. Contributed by Doug Nychka (nychka@stat.ncsu.edu). [2/Mar/95] (3 kbytes). The actual compressed tar file is available as slab.tar.Z in the S collection. Access this file via gopher, FTP, or the WWW, but not e-mail.
- sm
- The library `sm' is associated with the book `Applied Smoothing Techniques for Data Analysis: the Kernel Approach with S-Plus illustrations" by A.W.Bowman and A.Azzalini, 1997, Oxford University Press. A compressed tar file. This is version 2 of the library. Please send comments or questions to A.W.Bowman (adrian@statistics.glasgow.ac.uk) or A.Azzalini (adelchi@pearson.stat.unipd.it). There is also a MSwindows version available here [2/Jun/98][18/Dec/00] (191k)
- smoothers
- A set of S functions used to build a variety of Tukey median based smoothers and digital filters. Contributed by Colin Goodall (colin@jackknife.princeton.edu). [16/June/89](10664 bytes)
- spatial
- S interface to Fortran functions for universal kriging; K-fn and pseudo-likelihood analyses of spatial point patterns. Contains several spatial datasets. Superseded by Modapplstat.
- splines
- A shar achive of spline code for new S. It requires that you be able to dynamically load new C code. Contributed by Doug Bates (bates@stat.wisc.edu). [1/May/89],[11/Jan/90]{Revised 24/Jan/90} (15679 bytes)
- splusnotes
- Instructions for obtaining the LaTeX (and postscript) source, and associated data, for a short course on Splus. The document talks mostly about plain S features and t does not concentrate on features specific to Splus. Very useful as an introductory document. Supersedes snotes. Created by Bill Venables (wvenable@spam.ua.oz.au) [4/Aug/92][24/Mar/97]
- Spoetry
- pdf file containing a 429 page book on programming in (version 3 of) the S language. Also a shar file of accompanying functions, many of general utility. Submitted by Pat Burns pburns@pburns.seanet.com.
- spread
- An interactive spreadsheet interface to S/Splus. Submitted by, rmh@fisher.stat.temple.edu (Richard M. Heiberger). [25/Apr/96] (171k)
- sshc
- The function sshc() calculates sample size for historical control trials using the randomized trial formula, the Makuch-Simon method, and the uniform power method. Submitted by J. Jack Lee (jjlee@mdanderson.org). [10/Nov/00] (8k)
- str
- Compactly print STRucture of any S object. S function 'str', help, utilities and examples. Martin Maechler (maechler@stat.math.ethz.ch) [11/Mar/94] [29/Nov/94] [24/Mar/97] (35k)
- stukel
- Fits Stukel's generalization of the logistic link function to yes/no data. Shar archive. bwb@loki.mda.uth.tmc.edu (Barry W. Brown) [2/Oct/95] (246k)
- sunflower
- Sunflower plots, Martin Maechler (maechler@stat.math.ethz.ch) [1/Jun/95] (14 k)
- survcart
- S and C code for Classification And Regression Trees analysis with survival data. A shar archive. Georg Haupt, Ulrich Mansmann, (haupt@ukbf.fu-berlin.de) [1/Jun/95] (16 kbytes)
- survival
- Obsolete, see survival2 below.
- survival2
- Functions for survival analysis including cox regression, Kaplan-Meier estimationm Time-dependent and/or multiple event Cox models. A shar archive. (therneau@mayo.edu) [17/June/91] (270786 bytes)
- survival3
- Functions for survival analysis. Encompasses all of survival2, + uses the "models and methods" style of 8/91 S, plus several new functions. (therneau@mayo.edu) [20/July/92] [7/Aug/92] [14/Apr/93] (446k)
- survival3.doc
- Ancillary documentation for "survival3". A 30 page writeup of comments and examples (latex) plus some smaller notes on specific topics. (therneau@mayo.edu) [20/July/92] [7/Aug/92] [14/Apr/93] (153k)
- survival4
- Functions for survival analysis. A major update to survival3. (therneau@mayo.edu) [17/Oct/94][29/Nov/94] (596k) There is also a precompiled DOS/Windows version
- survival4.doc
- Ancillary documentation for "survival4", including a writeup of comments and examples (latex) plus some smaller notes on specific topics. (therneau@mayo.edu) [17/Oct/94] [29/Nov/94] (227k)
- surv5-34
- Survival5, including some updates not yet a part of S-Plus (depending on your release). For Splus 3.4. Terry Therneau (therneau@mayo.edu).[10/Jun/00] (1.7M)
- surv5
- Survival5, including some updates not yet a part of S-Plus (depending on your release). For Splus 5. Terry Therneau (therneau@mayo.edu).[10/Jun/00] (1.9M)
- s.to.latex
- Tools for converting S help files and S code to latex, and a corresponding latex .sty file. Submitted by Chambers and Hastie (trevor@research.att.com) [23/Mar/93] (15k)
- time.series
- A library of S functions for estimating, converting and analysing multivariate ARMA and state space models. Especially useful for testing statistics and estimation techniques and for examining small sample properties. The most current version of the archive, user guide, and additional information are also available at http://www.bank-banque-canada.ca/pgilbert and by anonymous ftp to ftp.bank-banque-canada.ca in the directory pub/pgilbert. [22/Oct/96] (shar file 1.19M bytes) P.Gilbert (pgilbert@bank-banque-canada.ca) See also for a Windows version.
- time.series.guide
- A postscript file with a brief user's guide for the archive time.series.[May/95] (file 250K bytes)
- timeslab
- A set of S functions and Fortran code for time-series analysis (modelled after the PC program of the same name). Includes documentation in TeX source. A shar archive. (jnewton@stat.tamu.end) [25/March/92] (164731 bytes).
- thresh
- S code to do one-dimensional data analytic wavelet threshold selection (must be used with "wavethresh" S-package by Nason and Silverman). Todd Ogden (ogden@math.scarolina.edu), [23/Jan/95] (10k)
- translate
- an S function that invokes unix() to use sed or tr to translate text strings. Part of the Harrell/hmisc collection.
- transcan
- Simultaneous transformation and imputation using canonical variates. Part of the Harrell/hmisc collection.
- tps
- A shar file with S functions and test datasets for logistic regression analysis of data collected using two phase stratified sampling. Submitted by Breslow and Chatterjee (nilanjan@stat.washington.edu) [9/Aug/98] (26858 bytes)
- transplot
- A shar file containing S and C functions for a graphical display to facilitate selecting a transformation of the response in an ANOVA or regression model. Submitted by Pat Burns (pat@statsci.com) [6/Feb/92] (20771 bytes)
- trax
- (Transform Plot Axes). Transform axes by a provided function as per built in log. Can thus do probit or logit plots, for example. Shar archive. bwb@loki.mda.uth.tmc.edu (Barry W. Brown) [2/Oct/95] (11k)
- tree
- A shar archive containing two functions to display S data objects in a tree-structured format. list.tree displays ordinary data objects (usually lists) and expr.tree displays expressions and functions. Contr. by (zaslavsk@dolphin.mit.edu) [24/May/89][7/Aug/91](8270 bytes)
- treefix
- A shar archive of functions to replace prune.tree and predict.tree, correcting errors and enhancing functionality. Both C-based and pure S versions are included. Part of the Modapplstat collection.
- tsgrowth
- An S-Function that fits the general linear growth curve model using the two-stage method when a linear combination of the growth curve coefficients is being modelled. Written by E. Demidenko and T.Stukel (eugene.demidenko@dartmouth.edu) [3/Jun/98] (9216 bytes)
- tssa
- A shar file containing S class, methods, and help files, a C engine, and postscript documention for Tree Structured Survival Analysis (TSSA) as described in Segal M.R. (1988), Regression trees for censored data, Biometrics, 44:35--47. Contributed by Mark Segal (mark@biostat.ucsf.edu) and Carrie Wager (cwager@hsph.harvard.edu). Version 1.16. [22/Oct/96](680K)
- tukey
- This shar archive Cumulative Probabilities (ptukey) and Quantiles (qtukey) of the Studentized Range, or (more generally) for the maximum of m studentized range statistics. Contributed by John Maindonald (john@maths.marc.cri.nz) [28/Aug/95] (36 Kb)
- tutor
- A menu-driven interactive tutorial program for Splus. It provides an introductory overview to many Splus functions and utilities. Contributed by Fu-Chuen Chang, (changfc@math.nsysu.edu.tw) [1/Aug/95] (29kbytes)
- uncond_f
- Exact tests for two independent binomial samples based on a wide range of non-classical Nullhypotheses and several families of test statistics. In data.dump format. Submitted by Ulrich Mansmann (mansmann@medizin.fu-berlin.de). [19/May/99] (67kbytes)
- varclus
- Hierarchical clustering of variables using various similarity measures (Pearson r^2, Spearman rho^2, Hoeffding D). Part of the Harrell/hmisc collection.
- vars
- List variables and selected attributes for a data frame or matrix. Submitted by Frank Harrell (fharrell@virginia.edu) [23/July/92] (5kbytes)
- visualizing.data
- This ascii file created by S contains 22 data sets from the book Visualizing Data published by Hobart Press (books@hobart.com) and written by William S. Cleveland (wsc@research.att.com). There is also a README object so there are 26 objects in all. The objects were written into the file by data.dump(), and should be restored by data.restore(). For example, if the dump is in the file xxx, the S command data.restore("xxx") reads them into S. The name of each data set is the name used in the book. To find the description of the data set, look under the entry "data, name" in the index. For example, one data set is barley. To find the description of barley, look in the index under the entry "data, barley". The datasets archive of Statlib also has the data in a table format, but, of course, it is far easier to use this dump to get the data into S. (671 kbytes) [12/Nov/93] [2/Oct/95] [22/Oct/97]
- visualizing.scripts
- This is a collection of S-scripts for producing figures in W.S.Cleveland's "Visualizing Data" book. Included is a README file and 270+ scripts. [17/Oct/94][1/Dec/95] [22/Oct/97] (128k)
- visualizing.tables
- This is a collection of tables from W.S.Cleveland's "Visualizing Data" book. [22/Oct/97] (534k)
- vied
- Enhanced vi and ed functions that allow multiple S objects to be edited and merged. Contributed by Colin Goodall (crg@stat.psu.edu). [6/Oct/93] (11k)
- vtsim
- S and Fortran 90 code for Nonparametric Regression Analysis of Longitudinal Data (Joan G. Staniswalis and J. Jack Lee, JASA, 93:1403-1418, 1998). A shar file. Submitted by J. Jack Lee (lee@odin.mdacc.tmc.edu).[13/Jan/99][14/May/99] (59k)
- wavecov
- A shar archive containing S functions in order to perform a wavelet analysis of variance on univariate and bivariate time series. Submitted by Brandon Whitcher (whitcher@eurandom.tue.nl). [13/Aug/98][5/Mar/99][27/Jul/99] (752k)
- wavethresh
- Code for doing wavelet transforms and thresholding in 1 and 2D in S. (G.P.Nason@maths.bath.ac.uk)[7/April/93] [Version 2.2, 7/July/93] (180kbytes)
- windows
- Window/kernel/moving-average smoothing/estimation /nonparametric regression. Central values, quantiles, derivatives. Shar file. Contr. by Barry Brown (an12354@mdaali.cancer.utexas.edu) [24/May/91] (153041 bytes)
- wlb
- S code to perform a weighted likelihood bootstrap. Michael Newton (newton@stat.wisc.edu) [17/Dec/92] (3k)
- world
- A simple S script for a world map. Submitted by Steve Wofsy, (scw@io.harward.edu) [17/Sept/91] (105019 bytes).
- world.hires
- Double resolution world map data. Shar archive, Submitted by Steve Wofsy, (scw@io.harward.edu) [1/Oct/91] (238558 bytes).
- 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]
- x11driver
- A shar archive containing an x11 driver for new S. Contributed by Mike Meyer (mikem@andrew.cmu.edu). [5/May/89](56629 bytes)

Last modified: Tue Dec 19 10:24:33 EST 2000 by Pantelis Vlachos