Software

  • Reproducible Science with LaTeX (with HaiYing Wang). Journal of Data Science. The LaTeX package runcode is available on CTAN.
  • The talk2stat Python package allows programs (such as latextopdf, nodejs) to communicate with statistical packages.
  • edgefinder (current version is 0.1.10) – a method for recovering a gene network’s structure from co-expression data. The code for simulations and a case study files are provided here edgefinder.tar. (PLOS ONE). The preferred way to install the package is directly from the github.com repository, by running
    devtools::install_github("haimbar/edgefinder")
    The zipped R package is also available here. To install from the tar.gz file, download the file and run install.packages("edgefinder_0.1.10.tar.gz",repos=NULL)  (Note that the web server may change the zipped files *tar.gz files to *tar_.gz, so rename the file by removing the extra underscore before running the install.packages command).
  • QREM – quantile regression via the EM algorithm, based on the paper on arxiv. The preferred way to install the package is directly from the github.com repository, by running
    devtools::install_github("haimbar/QREM")
    Or, you can download the R package QREM and install by using
    install.packages("QREM_0.1.8.tar.gz",repos=NULL) (Note that the web server changed the zipped files *tar.gz files to *tar_.gz, so rename the file by removing the extra underscore before running the install.packages command)
  • SEMMS – “Scalable EMpirical Bayes Model Selection” — a variable selection method which uses a parsimonious mixture model to identify significant predictors in the Generalized Linear Models (GLM) framework (JCGS 2019, arxiv).  Install the package from github, using
    devtools::install_github("haimbar/SEMMS"). The R package (version 0.2.5) contains some data sets, and a vignette with detailed examples. To install from the zip file, download the package and run install.packages("SEMMS_0.2.5.tar.gz",repos=NULL) (Note that the web server changed the zipped files *tar.gz files to *tar_.gz, so rename the file by removing the extra underscore before running the install.packages command). The project’s source is available on github at https://www.github.com/haimbar/SEMMS . Fifty free eprint versions of the paper are available from the publisher, Taylor and Francis. See also a short description in the YoungStatS blog.
  • DVX is an interactive program written in R, which can be used to perform Differential Variation and eXpression analysis of gene expression (or similar) data (STAT).
  • RCMap is an open-source concept mapping software. The latest version is now available on github. The older version is available here. For more background about concept mapping, its applications, and about R-CMap, see the Special Issue in Evaluation and Program Planning
  • MCMC code for the paper “A Spatial Capture-Recapture Model with Dependent Animal Movement” (in preparation), with Paul McLaughlin, Ph.D. Code was written by Dr. McLaughlin. To change the dataset, see the “User Input” section. The data has to include capture histories, trap locations, the number of trapping occasions, the size of a buffer zone around the trapping area, chain length and burn-in settings for the MCMC code, and the upper bound for the number of animals.