Department of Statistics Home


Software written by members of the Statistics Department and/or their collaborators available for download.

Software for Belief Networks
This document describes software for processing graphical belief function models, and related modes such as Bayesian networks, influence diagrams, and probabilistic graphical models. (Russell Almond)

Polynomial Splines
Software for density estimation (LOGSPLINE), hazard estimation (HEFT), hazard regression (HARE) and spectral density estimation (LSPEC) using polynomial splines. (Charles Kooperberg)

Software for Pedigree Analysis for Genetics (and Epidemiological Attributes). PANGAEA currently consists of four packages MORGAN, BOREL, PedPack, and HARDY. These subdirectories contain source-code of C programs for Genetic Analysis, including Monte Carlo segregation and linkage analysis (MORGAN), pedigree analysis and graphics (PedPack), and relationship inference (BOREL). Included are main programs, subroutine libraries, test data sets, and documentation. (Elizabeth Thompson)

Software for Bayesian Model Selection
Software for Bayesian hypothesis testing and model selection in linear regression, generalized linear models, survival analysis and other models. (Adrian Raftery and Chris Volinsky).

Model-Based Clustering Software
Software for model-based clustering of multivariate data, and for detecting features in spatial point processes. (Adrian Raftery and Chris Fraley).

Deterministic Simulation Models Software
Software for making inference from deterministic simulation models. The programs avaiable are specifically for whale population dynamics models, but could be extended to other applications. (Adrian Raftery and Geof Givens).

Monte Carlo Inference Software
Software for (1) determining the number of iterations in MCMC inference; (2) the weighted likelihood bootstrap (WLB); (3) MCMC algorithms for model uncertainty; and (4) using the sampling-importance-resampling (SIR) algorithm for inference about simulation models. (Adrian Raftery, Steven Lewis, Michael Newton, Jennifer Hoeting, and Geof Givens).

Time Series Software
Software for (1) estimating long memory (fractional differencing) time series models; (2) robust Bayesian estimation and order selection for autoregressive models; and (3) MTD models for higher-order Markov chains. (Adrian Raftery, Chris Fraley, Nhu Le, and Simon Tavare)

A set of S functions for classification and (polychotomous) regression using adaptively selected polynomial splines for the model building based on the MARS algorithm. (Charles Kooperberg and Martin O'Connor)