Planet R

CRANberries: New package sparsediscrim with initial version 0.1.0

2013-05-25T19:12:00+00:00

Package: sparsediscrim
Title: Sparse Discriminant Analysis
Version: 0.1.0
Date: 2013-05-25
Author: John A. Ramey
Maintainer: John A. Ramey
Description: A collection of sparse and regularized discriminant analysis methods intended for small-sample, high-dimensional data sets. The package includes the High-Dimensional Regularized Discriminant Analysis classifier. Furthermore, the package includes several Diagonal Discriminant Analysis methods, which are special cases of the well-known Naive Bayes classifier.
Imports: corpcor, bdsmatrix, mvtnorm
License: MIT
URL: https://github.com/ramey/sparsediscrim, http://ramhiser.com
Collate: 'dlda.r' 'dqda.r' 'sdlda.r' 'sdqda.r' 'stein-shrinkage.r' 'estimates.r' 'helper-intercept.r' 'data-intraclass.r' 'data-block-autocorrelation.r' 'tong-shrinkage.r' 'hdrda.r' 'helper-cov.r' 'helper-cv.r' 'helper.r'
Packaged: 2013-05-25 17:19:33 UTC; johnramey
NeedsCompilation: no
Repository: CRAN
Date/Publication: 2013-05-25 20:04:01

More information about sparsediscrim at CRAN

CRANberries: New package SDD with initial version 1.0

2013-05-25T15:12:00+00:00

Package: SDD
Type: Package
Title: Serial Dependence Diagrams
Version: 1.0
Date: 2013-02-17
Author: Luca Bagnato, Lucio De Capitani, Angelo Mazza and Antonio Punzo
Maintainer: Angelo Mazza
Description: This package allows for computing (and by default plotting) different types of serial dependence diagrams.
License: GPL-2
LazyLoad: yes
Depends: R (>= 2.15.0)
Imports: Hmisc,sm,tseries,stats,rpanel,rgl
Packaged: 2013-05-25 11:20:26 UTC; Angelo
NeedsCompilation: no
Repository: CRAN
Date/Publication: 2013-05-25 15:30:25

More information about SDD at CRAN

Revolutions: Because it's Friday: Game of Thrones Family Trees

David Smith — 2013-05-24T21:33:30+00:00

I'm loving the Game of Thrones series, but not having read the books it has been a bit difficult keeping up with all the various families, houses and their entanglements. The Game of Thrones Wiki has been a great resource to clear up my questions, but I wish I'd found Magdalena Maslowska's Illustrated Guide to Houses and Character Relationships sooner — it would have cleared up a lot of my confusion. Here's just a detail, centered on the Lannisters:

(Hat tip: Francisco Palm.) You can explore the details in the full-size version (click to enlarge), but beware spoilers if you're not caught up to Season 3 as it's showing in the US.

On the other hand, you could just work with the (slightly NSFW) Game of Thrones Cheat Sheet below the jump. It's not terribly helpful, but it captured pretty well how I had to refer to most of the characters for the first season and a half or so!

If you haven't started Game of Thrones yet, why not take a look at it this weekend? It's a great series rich in intrigue, politics and story: don't be turned off by the fantasy theme, which is actually a fairly light element (especially early on). Whatever you get up to, have a great weekend!

CRANberries: New package msgl with initial version 0.1.1

2013-05-24T21:12:00+00:00

Package: msgl
Type: Package
Title: High dimensional multiclass classification using sparse group lasso
Version: 0.1.1
Date: 2013-20-05
Author: Martin Vincent (Sparse group lasso), Conrad Sanderson (Armadillo), Jaakko Jarvi (Boost Tuple) and Jens Maurer (Boost Random)
Maintainer: Martin Vincent
Description: Sparse group lasso multiclass classification, suitable for high dimensional problems with many classes (see url). Fast algorithm for solving the multinomial sparse group lasso optimization problem. Use of multiple processors for cross validation and subsampling is supported through OpenMP. The Armadillo C++ library is used as the primary linear algebra engine. Armadillo is licensed under the MPL 2.0 (see url). The Armadillo C++ library is primarily developed at NICTA (Australia) by Conrad Sanderson, with contributions from around the world. Furthermore the package utilize various Boost libraries, in particular the Tuple library by Jaakko Jarvi and the Random library by Jens Maurer. The Boost libraries are licensed under the Boost Software License (see url).
URL: http://arxiv.org/abs/1205.1245 http://arma.sourceforge.net/, http://www.mozilla.org/MPL/2.0/, http://www.boost.org/, http://www.boost.org/users/license.html
License: GPL (>= 2)
LazyLoad: yes
Depends: R (>= 2.13.0), Matrix
Collate: 'msgl_multinomial.R'
Packaged: 2013-05-24 18:52:34 UTC; martin
NeedsCompilation: yes
Repository: CRAN
Date/Publication: 2013-05-24 21:40:44

More information about msgl at CRAN

Bioconductor Project Working Papers: A versatile test for equality of two survival functions based on weighted differences of Kaplan-Meier curves

2013-05-24T19:58:11+00:00

With censored event time observations, the logrank test is the most popular tool for testing the equality of two underlying survival distributions. Although this test is asymptotically distribution-free, it may not be powerful when the proportional hazards assumption is violated. Various other novel testing procedures have been proposed, which generally are derived by assuming a class of specific alternative hypotheses with respect to the hazard functions. The test considered by Pepe and Fleming (1989) is based on a linear combination of weighted differences of two Kaplan-Meier curves over time and is a natural tool to assess the difference of two survival functions directly. In this article, we take a similar approach, but choose weights which are proportional to the observed standardized difference of the estimated survival curves at each time point. The new proposal automatically makes weighting adjustments empirically. The new test statistic is aimed at a one-sided general alternative hypothesis, and is distributed with a short right tail under the null hypothesis, but with a heavy tail under the alternative. The results from extensive numerical studies demonstrate that the new procedure performs well under various general alternatives. The survival data from a recent cancer comparative study are utilized for illustrating the implementation of the process.

Revolutions: Sentiment analysis finds trouble in the Enron emails

David Smith — 2013-05-24T19:10:45+00:00

The Enron email dataset, collected during the FERC investigation of the Enron financial scandal, represents the largest publicly available set of emails. This makes theman ideal testbed for sentiment analysis algorithms. Ikanow's Andrew Strite used the open-source Infinit.e framework and a Hadoop cluster to generate sentiment scores for all of the Enron emails, and then used R to manipulate and analyze the resulting data. Here's a visualization of just a few of the email accounts: the red marks flag emails where the sender's sentiment suddenly turned sharply negative (and would therefore be a good place to start looking for evidence):

Andrew used the rjson package to interface with the Ikanow REST API, the plyr package to restructure the incoming data, and the ggplot2 package to visualize the results. In a subsequent analysis he also used the zoo package to interpolate and analyze time series of sentiment scores, which you can read about in the full blog post below.

Ikanow blog: Making the most of sentiment scores using Ikanow and R

CRANberries: New package tlnise with initial version 2.0

2013-05-24T17:12:00+00:00

Package: tlnise
Depends: R (>= 3.0.0)
Suggests: MASS
Imports: stats
Version: 2.0
Date: 2013-05-24
Title: Two-level normal independent sampling estimation
Author: S-PLUS original by Phil Everson; R port by Roger D. Peng
Maintainer: Roger D. Peng
Description: Functions for two level normal models as described in Everson and Morris (2000). J. R. Statist. Soc. B, 62 prt 2, pp.399--412.
License: GPL-3
URL: http://www.swarthmore.edu/NatSci/peverso1/TLNise/tlnise.htm
Packaged: 2013-05-24 14:37:36 UTC; rdpeng
NeedsCompilation: yes
Repository: CRAN
Date/Publication: 2013-05-24 18:24:21

More information about tlnise at CRAN

CRANberries: New package SparseTSCGM with initial version 1.0

2013-05-24T15:12:00+00:00

Package: SparseTSCGM
Type: Package
Title: Sparse time series chain graphical models
Version: 1.0
Date: 2013-05-24
Author: Fentaw Abegaz and Ernst Wit
Maintainer: Fentaw Abegaz
Depends: R (>= 3.0.0), glasso, longitudinal, flare, mvtnorm, network
Description: Compute and plot sparse autoregressive coefficient and precision matrices for time series chain graphical models.
License: GPL (>= 3)
LazyLoad: yes
Packaged: 2013-05-24 12:05:56 UTC; P261204
NeedsCompilation: yes
Repository: CRAN
Date/Publication: 2013-05-24 15:38:46

More information about SparseTSCGM at CRAN

CRANberries: New package compositionsGUI with initial version 1.0

2013-05-24T15:12:00+00:00

Package: compositionsGUI
Type: Package
Title: Graphical User Environment for Compositional Data Analysis
Version: 1.0
Date: 2013-05-14
Author: Jiri Eichler, Karel Hron, Raimon Tolosana-Delgado, Gerald van den Boogaart, Matthias Templ, Peter Filzmoser
Maintainer: Jiri Eichler
Depends: robCompositions, compositions, mvoutlier, robustbase, RGtk2, MASS, nnet, boot
Description: The package provides graphical user interface and teaching tool for compositional data analysis using R. It contains data loading from various sources (external sources, installed R packages), preprocessing (imputation of missing values and rounded zeros), basic summary statistics as well as log-ratio transformations, necessary for a reasonable statistical treatment of compositional data (isometric, centred, additive log-ratio transformations). Further, also statistical methods like outlier detection and principal component analysis, together with the corresponding graphical tools, are provided. Finally, the package contains plotting functions, suitable for compositional data analysis, like ternary and quaternary diagrams, scatterplots of log-ratio transformed data, and diagrams with categorical/continuous covariates.
License: GPL (>= 2)
Packaged: 2013-05-16 20:20:57 UTC; templ
NeedsCompilation: no
Repository: CRAN
Date/Publication: 2013-05-24 12:34:41

More information about compositionsGUI at CRAN

CRANberries: New package Demerelate with initial version 0.8-0

2013-05-24T09:12:00+00:00

Package: Demerelate
Version: 0.8-0
Date: 2012-05-24
Title: Functions to calculate relatedness on diploid genetic data
Authors@R: c(person("Philipp","Kraemer",role=c("aut", "cre"),email="Philipp.Kraemer@uni-oldenburg.de"),person("Gabriele","Gerlach",role="aut",email="Gabriele.Gerlach@uni-oldenburg.de"))
Author: Philipp Kraemer and Gabriele Gerlach
Maintainer: Philipp Kraemer
Depends: R (>= 2.15.0), fts, mlogit, sfsmisc
Suggests: MASS
Description: Functions to calculate pairwise relatedness on diploid genetic datasets. Different estimators for relatedness can be combined with information on geographical distances. Information on heterozygosity, allele- and genotype diversity as well as genetic F-statistics are provided for each population.
License: GPL (>= 2)
URL: http://www.r-project.org
BugReports: http://Demerelate.bugtracker.url
Packaged: 2013-05-24 08:10:41 UTC; philipp
NeedsCompilation: no
Repository: CRAN
Date/Publication: 2013-05-24 10:29:00

More information about Demerelate at CRAN

CRANberries: New package twostageTE with initial version 1.0

2013-05-24T07:12:00+00:00

Package: twostageTE
Type: Package
Title: Two-Stage Threshold Estimation
Version: 1.0
Date: 2013-03-05
Author: Shawn Mankad, George Michailidis, Moulinath Banerjee
Maintainer: Shawn Mankad
Description: Implements a variety of nonparametric methods for computing one-stage and two-stage confidence intervals, as well as point estimates of threshold values.
License: GPL-2
Depends: isotone
Packaged: 2013-05-23 22:28:03 UTC; smankad
NeedsCompilation: no
Repository: CRAN
Date/Publication: 2013-05-24 07:38:34

More information about twostageTE at CRAN

Revolutions: Highlights of the Milwaukee Workshop on R and Bioinformatics

Joseph Rickert — 2013-05-23T17:41:49+00:00

by Joseph Rickert

On May 10th and 11th, in honor of this being the International Year of Statistics, the Milwaukee Chapter of the American Statistical Association (MILWASA) held a workshop on cutting edge uses of R in Bioinformatics. One objective of the workshop was to show the "nuts and bolts" details of how R with C++ integration and the specialized capabilities of the Bioconductor Project provides an flexible, feature-rich platform for advanced Bioinfomatics applications. Featured speakers were:

Denise Scholtens who gave talks on analyzing microarray data using R and Bioconductor, building graphs with R and Bioconductor, gene set enrichment analysis, and Expression Set objects.
Kwang-Youn Kim who spoke on the analysis of RNA sequencing data using R and Bioconductor and
Dirk Eddelbuettel gave a thorough, four-part introduction to Rcpp, his package for integrating R with C++.

A tremendous amount of material from this workshop (pdfs, slides, data and R code) is available online. And, if you are interested in R and C++ integration have a look at Dirk’s new book.

The following graph from Denise’s presentation on gene set enrichment analysis shows a portion of an induced gene ontology graph using using the classic Fisher elimination algorithm and gives an idea of the some of the sophisticated analyses you can do with her R code.

Note if you want to run the code you will have to get some of the packages from Bioconductor. Here is some code from Kwang-Youn on how to get started.

# Install all the necessary packages if not on your system
source("http://bioconductor.org/biocLite.R")
## Bioconductor version 2.11 (BiocInstaller 1.8.3), ?biocLite for help
biocLite(c("TxDb.Dmelanogaster.UCSC.dm3.ensGene", "ShortRead", "edgeR", "cummeRbund"))
## BioC mirror: http://bioconductor.org
## Using Bioconductor version 2.11 (BiocInstaller 1.8.3), R version 2.15.
## Installing package(s) 'TxDb.Dmelanogaster.UCSC.dm3.ensGene' 'ShortRead'
## 'edgeR' 'cummeRbund'
##
## The downloaded binary packages are in
## /var/folders/nk/9bnzzk_152vg4wslcbxc5g_c0000gn/T//RtmpzkrvEW/downloaded_packages

Created by Pretty R at inside-R.org

And here is some sample code from Dirk's presentation on calling R plot functions from C++.

#include <RInside.h> // embedded R via RInside
int main(int argc, char *argv[]) {
RInside R(argc, argv); // create an embedded R instance
// evaluate an R expression with curve()
std::string cmd = "tmpf <- tempfile(’curve’); "
"png(tmpf); curve(x^2, -10, 10, 200); "
"dev.off(); tmpf";
// by running parseEval, we get ﬁlename back
std::string tmpfile = R.parseEval(cmd);
std::cout << "Could use plot in " << tmpfile << std::endl;
unlink(tmpfile.c_str()); // cleaning up
// alternatively, by forcing a display we can plot to screen
cmd = "x11(); curve(x^2, -10, 10, 200); Sys.sleep(30);";
R.parseEvalQ(cmd);
exit(0);
}

Revolution Analytics is proud to have been a sponsor for this workshop. Congratulations to Rodney Sparapani of the Medical College of Wisconsin for making it happen!

CRANberries: New package InferenceSMR with initial version 1.0

2013-05-23T17:12:00+00:00

Package: InferenceSMR
Type: Package
Title: Inference about the standardized mortality ratio when evaluating the effect of a screening program on survival.
Version: 1.0
Date: 2013-05-22
Author: Denis Talbot, Thierry Duchesne, Jacques Brisson, Nathalie Vandal
Maintainer: Denis Talbot
Description: The InferenceSMR package provides functions to make inference about the standardized mortality ratio (SMR) when evaluating the effect of a screening program. The package is based on methods described in Sasieni (2003) and Talbot et al. (2011).
License: GPL (>= 2)
Depends: survival
Packaged: 2013-05-23 14:10:09 UTC; Nilidas
NeedsCompilation: no
Repository: CRAN
Date/Publication: 2013-05-23 17:41:26

More information about InferenceSMR at CRAN

Revolutions: 7th R/Rmetrics workshop in Switzerland, June 30-July 4

David Smith — 2013-05-23T16:33:49+00:00

The 7th annual R/Rmetrics Workshop om Computational Finance and Financial Engineering will take place June 30-July 4 in the beatiful alpine setting of Lake Thune, Switzerland. This is an intimate workshop limited to around 50 participants, and features tutorials from leading practitioners in finance with R, with a special focus on the Rmetrics suite of R packages. This year's program includes in-depth material from experts in academia and the finance industry, as you can see below:

Key Note Speaker:

Gunter Loeffler - University of Ulm, Institute of Finance
    Tower Building and Stock Market Returns

Tutorials:

Basics and Fundamentals:
Nicolas Polson, University of Chicago, School of Business, USA
     Bayesian Inference, Gibbs Sampling and Markov Chain Monte Carlo
Stefano Iacus, University of Milano, Department of Economics and Statistics, Milano, Italy
    Quasi Likelihood Inference and Model Selection for Stochastic Differential Equations

Modern Portfolio Design:
Bernhard Pfaff, Invesco Research Frankfurt, Germany
    Portfolio Selection, Optimization and Design with R
Diethelm Wuertz, Swiss Federal Institute of Technology, Zurich, Switzerland
    Portfolio Diversification and Stability Strategies

Advanced Computing in R:
Stefan Theussl, Raiffeisen Research and Vienna Univeristy of Economics, Austria
    High Performance Computing and Parallel R

Building Platforms:
Charles Roosen, Zurich Re Insurance, Zurich, Switzerland
    Behind the Zurich Re Insurance Platform
Wolfgang Breymann, Zurich University of Applied Sciences, Switzerland
    The Unified Financial Modeling Platform

For more information about the workshop, follow the link below.

Rmetrics.org: 2013 Meielisalp Workshop and Summer School

Bioconductor Project Working Papers: Subsemble: An Ensemble Method for Combining Subset-Specific Algorithm Fits

2013-05-23T16:17:02+00:00

Ensemble methods using the same underlying algorithm trained on different subsets of observations have recently received increased attention as practical prediction tools for massive datasets. We propose Subsemble: a general subset ensemble prediction method, which can be used for small, moderate, or large datasets. Subsemble partitions the full dataset into subsets of observations, fits a specified underlying algorithm on each subset, and uses a clever form of V-fold cross-validation to output a prediction function that combines the subset-specific fits. We give an oracle result that provides a theoretical performance guarantee for Subsemble. Through simulations, we demonstrate that Subsemble can be a beneficial tool for small to moderate sized datasets, and often has better prediction performance than the underlying algorithm fit just once on the full dataset. We also describe how to include Subsemble as a candidate in a SuperLearner library, providing a practical way to evaluate the performance of Subsemble relative to the underlying algorithm fit just once on the full dataset.

CRANberries: New package SODC with initial version 1.0

2013-05-23T15:12:00+00:00

Package: SODC
Type: Package
Title: Optimal Discriminant Clustering(ODC) and Sparse Optimal Discriminant Clustering(SODC)
Version: 1.0
Date: 2013-05-15
Author: Yanhong Wang
Maintainer: Yanhong Wang
Description: To implement two clustering methods, ODC and SODC, for clustering datasets using optimal scoring, can also be used as an dimension reduction tool.
License: GPL-2
Depends: magic,ppls,psych,MASS
Packaged: 2013-05-23 13:34:01 UTC; matyxwx
NeedsCompilation: no
Repository: CRAN
Date/Publication: 2013-05-23 15:53:51

More information about SODC at CRAN

Revolutions: Vote in the KDnuggets poll on Analytics Software

David Smith — 2013-05-22T22:22:20+00:00

The 14th annual KDnuggets poll measuring use of analytics software is open for voting. The poll asks, "What Predictive Analytics, Big Data, Data mining, Data Science software you used in the past 12 months for a real project?" and allows up to 20 choices from commercial software, open source software, and "big data" software. R was the leading choice in 2012, and currently leads the pack in the voting for 2013.

You can enter your vote now at the link below.

KDnuggets: New Poll: Predictive Analytics, Big Data, Data Mining, Data Science Software Used

CRANberries: New package SnowballC with initial version 0.5

2013-05-22T19:12:00+00:00

Package: SnowballC
Type: Package
Version: 0.5
Date: 2013-05-22
Title: Snowball stemmers based on the C libstemmer UTF-8 library
Authors@R: person("Milan", "Bouchet-Valat", email="nalimilan@club.fr", role=c("aut", "cre"))
Description: An R interface to the C libstemmer library that implements Porter's word stemming algorithm for collapsing words to a common root to aid comparison of vocabulary. Currently supported languages are Danish, Dutch, English, Finnish, French, German, Hungarian, Italian, Norwegian, Portuguese, Romanian, Russian, Spanish, Swedish and Turkish.
License: BSD
Copyright: Dr Martin Porter (2001) for the libstemmer C library, and Milan Bouchet-Valat (2013) for the R package contents
URL: https://r-forge.r-project.org/projects/r-temis/
BugReports: https://r-forge.r-project.org/tracker/?group_id=1437
Packaged: 2013-05-22 12:55:04 UTC; milan
Author: Milan Bouchet-Valat [aut, cre]
Maintainer: Milan Bouchet-Valat
NeedsCompilation: yes
Repository: CRAN
Date/Publication: 2013-05-22 18:27:08

More information about SnowballC at CRAN

CRANberries: New package epibase with initial version 0.1-0

2013-05-22T19:12:00+00:00

Package: epibase
Version: 0.1-0
Date: 2013/01/11
Title: basic tools for the analysis of disease outbreaks.
Author: {The Hackout team (In alphabetic order: David Aanensen, Marc Baguelin, Paul Birrell, Simon Cauchemez, Anton Camacho, Caroline Colijn, Anne Cori, Xavier Didelot, Ken Eames, Christophe Fraser, Simon Frost, Niel Hens, Joseph Hugues, Thibaut Jombart, Lulla Opatowski, Oliver Ratmann, Samuel Soubeyrand, Marc Suchard, Jacco Wallinga, Rolf Ypma)}
Maintainer: Thibaut Jombart
Suggests: EpiEstim
Depends: R (>= 2.10), methods, MASS, ape, adegenet, network, networkDynamic, ggplot2, ggmap, sna, scales, plyr, reshape2, mapproj, RColorBrewer, knitr
Description: basic tools for the analysis of disease outbreaks.
License: GPL (>= 2)
LazyLoad: yes
VignetteBuilder: knitr
Collate: 'auxClasses.R' 'generics.R' 'obkContacts.R' 'obkSequences.R' 'obkData.R' 'obkData_Accessors.R' 'subset.R' 'obkData_basics.R' 'uniqSequences.R' 'dna2uniqSequences.R' 'make.phylo.R' 'phylo2ggphy.R' 'plotggMST.R' 'plotggphy.R' 'plot.R' 'plotIndividualTimeline.R' 'simuEpi.R' 'utils.R' 'plotGeo.R' 'annotatedTreeReader.R' 'zzz.R' 'epibase.R'
Packaged: 2013-05-22 15:09:24 UTC; thibaut
NeedsCompilation: no
Repository: CRAN
Date/Publication: 2013-05-22 18:34:00

More information about epibase at CRAN

Revolutions: R/Finance 2013 - quick take

Joseph Rickert — 2013-05-22T18:56:27+00:00

by Joseph Rickert

Last Friday and Saturday the fifth annual R/Finance Conference was held in Chicago: two coffee-fueled days of stochastic models, mostly very high quality presentations and most excellent conversation. R/Finance is a small conference. Attendance this year was capped at 300. My very rough informal estimate is that 30% of the attendees were academics, 10% vendors selling finance products and the rest either quants or quantitative traders with insatiable appetites for coding up mathematical theory that might provide a competitive edge. The organizers should be commended for maintaining the quality and excitement that characterized the first years of the conference and for running a well thought-out and nearly flawless event. In my experience, it is rare for a comnference to sustain the "buzz" that characterized its early years.

At the time I'm writing, the presentations have not yet been uploaded to the R/Finance site. However, those of you seeking an early look at some what went on can find the the slides from Michael Kapler's lightning talk on Portfolio Allocation with Cluster Risk Parity, Mathew Dowle's talk on data.table, and Dirk Eddelbuettel's talk RcppArmadillo: Accelerating R with C++ Linear Algebra.

This was the third time I was able to attend the event, but Revolution Analytics is proud to have been a sponsor of R/Finance from the beginning.

I am already looking forward to R/Finance 2014, but I hope to have more to say about this year's event later.

Bioconductor Project Working Papers: Targeted Maximum Likelihood Estimation for Dynamic and Static Longitudinal Marginal Structural Working Models

2013-05-22T16:43:49+00:00

This paper presents a novel targeted maximum likelihood estimator (TMLE) estimator for the parameters of longitudinal static and dynamic marginal structural models.We consider a longitudinal data structure consisting of baseline covariates, time-dependent intervention nodes, intermediate time-dependent covariates, and a possibly time dependent outcome. The intervention nodes at each time point can include a binary treatment as well as a right-censoring indicator. Given a class of dynamic or static interventions, a marginal structural model is used to model the mean of the intervention specific counterfactual outcome as a function of the intervention and time point.Because the true shape of this function is rarely known, the marginal structural model is used as a working model. The causal quantity of interest is defined as the projection of the true function onto this working model. We introduce a new pooled TMLE for the parameters of such marginal structural working models, and compare this estimator to a recently proposed stratified TMLE that is based on estimating the intervention-specific mean separately for each intervention of interest. The performance of the pooled TMLE is compared to the performance of the stratified TMLE and the performance of inverse probability weighted estimators using simulations. Concepts are illustrated using an example in which the aim is to estimate the causal effect of delayed switch following immunological failure of first line antiretroviral therapy among HIV infected patients. Data from the International epidemiological Databases to Evaluate AIDS, Southern Africa are analyzed to investigate this question using both TMLE and IPW estimators. Our results demonstrate practical advantages over an IPW estimator for working marginal structural models for survival, as well as cases in which the pooled TMLE is superior to its stratified counterpart.

CRANberries: New package BSquare with initial version 1.0

2013-05-22T15:12:00+00:00

Package: BSquare
Type: Package
Title: Bayesian Simultaneous Quantile Regression
Version: 1.0
Date: 2013-05-04
Author: Luke Smith & Brian Reich
Maintainer: Luke Smith
Depends: quadprog, quantreg, VGAM
Suggests: survival
Description: This package models the quantile process as a function of predictors.
License: GPL (>= 2)
Packaged: 2013-05-23 15:53:32 UTC; smithinger
NeedsCompilation: yes
Repository: CRAN
Date/Publication: 2013-05-22 17:07:20

More information about BSquare at CRAN

CRANberries: New package trueskill with initial version 0.1

2013-05-22T07:12:00+00:00

Package: trueskill
Title: Implementation the TrueSkill algorithm in R
Description: An implementation of the TrueSkill algorithm (Herbrich, R., Minka, T. and Grapel, T) in R; a Bayesian skill rating system with inference by approximate message passing on a factor graph. Used by Xbox to rank gamers and identify appropriate matches. http://research.microsoft.com/en-us/projects/trueskill/default.aspx Current version allows for one player per team. Will update as time permits. Requires R version 3.0 as it is written with Reference Classes. URL: https://github.com/bhoung/trueskill-in-r Acknowledgements to Doug Zongker and Heungsub Lee for their python implementations of the algorithm and for the liberal reuse of Doug's code comments (@dougz and @sublee on github).
Version: 0.1
Author: Brendan Houng
Maintainer: Brendan Houng
License: GPL-3
Depends: R (>= 3.0)
Imports: methods
Collate: 'factorgraph.r' 'init.r' 'competition.r' 'player.r'
Packaged: 2013-05-22 05:32:15 UTC; brendanhoung
NeedsCompilation: no
Repository: CRAN
Date/Publication: 2013-05-22 08:00:08

More information about trueskill at CRAN

CRANberries: New package support.BWS with initial version 0.1-0

2013-05-22T07:12:00+00:00

Package: support.BWS
Type: Package
Title: Basic functions for supporting an implementation of best-worst scaling
Version: 0.1-0
Date: 2013-05-21
Author: Hideo Aizaki
Maintainer: Hideo Aizaki
Description: This package provides three basic functions that support an implementation of object case (Case 1) best-worst scaling: one for converting a two-level orthogonal main-effect design/balanced incomplete block design into questions; one for creating a data set suitable for analysis; and one for calculating count-based scores.
License: GPL (>= 2)
Suggests: DoE.base, crossdes, survival
Packaged: 2013-05-21 23:49:34 UTC; user
NeedsCompilation: no
Repository: CRAN
Date/Publication: 2013-05-22 07:47:15

More information about support.BWS at CRAN

CRANberries: New package OPDOE with initial version 1.0-6

2013-05-22T07:12:00+00:00

Package: OPDOE
Version: 1.0-6
Title: OPtimal Design Of Experiments
Author: Petr Simecek , Juergen Pilz Mingui Wang , Albrecht Gebhardt .
Maintainer: Albrecht Gebhardt
Description: Experimental Design
Depends: orthopolynom, conf.design, mvtnorm, nlme, gmp, crossdes
License: GPL (>= 2)
Packaged: 2013-05-22 05:19:35 UTC; alge
NeedsCompilation: no
Repository: CRAN
Date/Publication: 2013-05-22 07:53:44

More information about OPDOE at CRAN

CRANberries: New package cyphid with initial version 1.1

2013-05-22T07:12:00+00:00

Package: cyphid
Type: Package
Title: Cycle and Phase Identification for mastication data
Version: 1.1
Date: 2013-04-04
Author: Elizabeth Crane
Maintainer: Elizabeth Crane
Depends: fda
Description: This library contains a primary function that divides chewing sequences in cycles and cycles into phases. See get.all.breaks for an example.
License: GPL (>= 2)
LazyLoad: yes
LazyData: yes
Packaged: 2013-05-22 02:56:31 UTC; bcrane
NeedsCompilation: no
Repository: CRAN
Date/Publication: 2013-05-22 07:52:43

More information about cyphid at CRAN

Revolutions: Big Data Republic's list of Top 20 Big Data Accounts on Twitter

David Smith — 2013-05-22T00:19:55+00:00

I'm very proud to have been selected today for Big Data Republic's list of the Top 20 Big Data Twitter accounts. With such Big Data practitioners, analysts and thought leaders like Kirk Borne, Hilary Mason, Pete Skomoroch, Mike Olson and Doug Henshchen on the list, I'm humbled to be listed at #3. My co-presenters from the recent data science webinar, Carla Gentry and Gregory Piatetsky (#1) are also on the list.

Special thanks to judging panel member (and Teradata's Chief Analytics Officer) Bill Franks, for these kind words about my tweets:

He seems to come up with content on a more frequent basis than most. At the same time, his content is consistently good. That’s a hard combination to find.

If you'd like to follow all 20 of the accounts recognized by the judging panel, I created this Twitter list: BDR Top 20 #BigData. Check out at all of the members and the comments from the judging panel at the link below.

Big Data Republic: @kdnuggets Voted the Best Big Data Twitter Account!

Revolutions: Get your questions answered about Open Data

David Smith — 2013-05-21T23:34:15+00:00

The OpenData StackExchange site has just launched in beta, and looks to be a great resource for open data sources. Like StackOverflow for programming and CrossValidated for statistics, OpenData is is a question and answer site for developers and researchers interested in open data. There's no R tag yet (though that would be nice for data sources specifically compatible with R), but there are already some useful tags for government data, APIs, and tools (to name just a few). If you have expertise on open data to share, or just have a question you need an answer to, follow the link below and check it out.

StackExchange: Open Data beta (via JM)

Bioconductor Project Working Papers: VARYING INDEX COEFFICIENT MODELS

2013-05-21T21:59:50+00:00

It has been a long history of utilizing interactions in regression analysis to investigate interactive effects of covariates on response variables. In this paper we aim to address two kinds of new challenges resulted from the inclusion of such high-order effects in the regression model for complex data. The first kind arises from a situation where interaction effects of individual covariates are weak but those of combined covariates are strong, and the other kind pertains to the presence of nonlinear interactive effects. Generalizing the single index coefficient regression model (Xia and Li, 1999), we propose a new class of semiparametric models with varying index coefficients, which enables us to model and assess nonlinear interaction effects between grouped covariates on the response variable. As a result, most of the existing semiparametric regression models are special cases of our proposed models. We develop a numerically stable and computationally fast estimation procedure utilizing both profile least squares method and local fitting. We establish both estimation consistency and asymptotic normality for the proposed estimators of index coefficients as well as the oracle property for the nonparametric function estimator. In addition, a generalized likelihood ratio test is provided to test for the existence of interaction effects or the existence of nonlinear interaction effects. Our models and estimation methods are illustrated by both simulation studies and an analysis of body fat dataset.

Bioconductor Project Working Papers: Surrogacy Assessment Using Principal Stratification When Surrogate and Outcome Measures are Multivariate Normal

2013-05-21T21:59:44+00:00

CRANberries: New package SphericalCubature with initial version 1.0

2013-05-21T21:12:00+00:00

Package: SphericalCubature
Type: Package
Title: Numerical integration over spheres and balls in n-dimensions; multivariate polar coordinates
Version: 1.0
Date: 2013-05-16
Author: John P. Nolan, American University
Maintainer: John P. Nolan
Depends: R (>= 2.1.15), cubature
Description: This package defines several methods to integrate functions over the unit sphere and ball in n-dimensional Euclidean space. Routines for converting to/from multivariate polar/spherical coordinates are also provided.
License: GPL (>= 2)
Packaged: 2013-05-21 17:56:00 UTC; jpnolan
NeedsCompilation: no
Repository: CRAN
Date/Publication: 2013-05-21 22:45:58

More information about SphericalCubature at CRAN

CRANberries: New package hint with initial version 0.1-0

2013-05-21T15:12:00+00:00

Package: hint
Type: Package
Title: Tools for hypothesis testing based on Hypergeometric Intersection distributions
Version: 0.1-0
Date: 2013-05-16
Author: Alex T. Kalinka
Maintainer: Alex T. Kalinka
Description: Hypergeometric Intersection distributions are a broad group of distributions that describe the probability of picking intersections when drawing independently from two (or more) urns containing variable numbers of balls belonging to the same n categories.
License: GPL (>= 2)
LazyLoad: yes
Packaged: 2013-05-21 12:17:26 UTC; alexk
NeedsCompilation: yes
Repository: CRAN
Date/Publication: 2013-05-21 16:51:44

More information about hint at CRAN

CRANberries: New package gazetools with initial version 3.0

2013-05-21T15:12:00+00:00

Package: gazetools
Type: Package
Title: Gaze Tools
Version: 3.0
Date: 2013-05-21
Author: Ryan M. Hope
Maintainer: Ryan M. Hope
URL: http://ryanhope.github.io/gazetools
BugReports: https://github.com/RyanHope/gazetools/issues
Description: A collection of tools for processing and classifying eye gaze data.
License: GPL-3
LazyData: true
Enhances: sp, ggplot2
Depends: R (>= 2.14), Matrix, methods,
Imports: rgeos, maptools, deldir, sp, modeest, signal, zoo, Rmisc, reshape, plyr, spatstat, ggplot2
Collate: 'event_ids.R' 'distance2point.R' 'subtendedAngle.R' 'angle2pixels.R' 'local_maxima.R' 'find_peak_ranges.R' 'find_peaks.R' 'pva.R' 'classify.R' 'classify.V.R' 'classify.VA.R' 'classify.VI.R' 'mould.R' 'mouldThreshold.R' 'voronoi.R' 'chull_area.R' 'nni.R' 'bcea.R' 'gridded_rois.R' 'roi_coverage.R' 'gazetools.R' 'extent.R' 'pva-methods.R' 'classify-methods.R' 'mould-methods.R'
Packaged: 2013-05-21 14:11:59 UTC; ryanhope
NeedsCompilation: no
Repository: CRAN
Date/Publication: 2013-05-21 16:53:01

More information about gazetools at CRAN

CRANberries: New package SetMethods with initial version 1.0

2013-05-21T13:12:00+00:00

Package: SetMethods
Type: Package
Title: SetMethods: A Package Companion to "Set-Theoretic Methods for the Social Sciences".
Version: 1.0
Date: 2013-05-17
Author: Mario Quaranta
Maintainer: Mario Quaranta
Description: SetMethods is a package companion to the book by C. Q. Schneider and C. Wagemann "Set-Theoretic Methods for the Social Sciences", Cambridge University Press. It contains some additional functions not present in other packages and data to replicate the examples in the book and in the online appendix.
Depends: lattice, betareg, R (>= 2.14.0)
Imports: lattice, betareg
License: GPL-2
Packaged: 2013-05-21 09:50:30 UTC; mario
NeedsCompilation: no
Repository: CRAN
Date/Publication: 2013-05-21 14:52:42

More information about SetMethods at CRAN

CRANberries: New package overlap with initial version 0.1.0

2013-05-21T07:12:00+00:00

Package: overlap
Type: Package
Title: Estimates of coefficient of overlapping for animal activity patterns
Version: 0.1.0
Date: 2013-05-21
Author: Mike Meredith and Martin Ridout
Maintainer: Mike Meredith
Description: Provides functions to fit kernel density functions to data on temporal activity patterns of animals; estimate coefficients of overlapping of densities for two species; and calculate bootstrap estimates of confidence intervals.
License: GPL (>= 3)
Packaged: 2013-05-21 04:06:09 UTC; MIKE
NeedsCompilation: yes
Repository: CRAN
Date/Publication: 2013-05-21 07:18:59

More information about overlap at CRAN

Revolutions: R programming challenge: Escape the zombie horde

David Smith — 2013-05-20T21:42:01+00:00

So when the world is taken over by a Zombie horde, you're going to want to figure out a way to get the human population to safety. This R script by econometrician Francis Smart won't help you do that exactly, but given a list of waypoints to navigate through zombie-infested lands to a safe house, it will tell you how many how many members of your human party survive:

While zombie-escaping might not seem to be a terribly useful application of R (until the outbreak comes, anyway), Francis's script comes with detailed comments that make it a useful example of simulation and animations for budding R programmers. (Try switching up the seed used in the set.seed call to get a new horde of zombies to play with, and experiment with new values of the waypoints variable to guide your hapless humans to safety.) It's also a great example of using the animation package to crate animated GIFs with R.

If you're looking for a programming challenge, an interesting experiment would be to write a function to automatically select waypoints on the route to safety. You could then run the simulation over and over (using a random seed, this time) and count the number of survivors to evaluate its effectiveness. Let us know what you come up with in the comments.

Econometrics by Simulation: Strategic Zombie Simulation - Animation - Jimmy's Down

CRANberries: New package nloptwrap with initial version 0.5-1

2013-05-20T21:12:00+00:00

Package: nloptwrap
Type: Package
Title: Wrapper for Package nloptr
Version: 0.5-1
Date: 2013-05-09
Author: Hans W Borchers
Maintainer: HwB
Depends: R (>= 2.11.1), nloptr
Description: Wraps functions in 'nloptr' for easier access and use.
License: LGPL
LazyLoad: yes
LazyData: yes
Repository: CRAN
Repository/R-Forge/Project: optimist
Repository/R-Forge/Revision: 300
Repository/R-Forge/DateTimeStamp: 2013-05-09 15:38:19
Date/Publication: 2013-05-20 22:26:16
Packaged: 2013-05-09 18:15:54 UTC; rforge
NeedsCompilation: no

More information about nloptwrap at CRAN

Simon Jackman: swings in the Australian election betting markets

2013-05-20T19:12:02+00:00

A little bit of volatility in the Federal election betting markets to report…

Over the last 12 days Labor’s price at Centrebet has recovered from $9.60 to $6.50, the Coalition out from $1.05 to $1.10. The Labor win probabilities implied by these prices are 9.8% and 14.5%, so in relative terms this is a reasonably big swing; indeed, the rate of return on a successful Coalition wager has doubled (5% to 10%).

Labor had been on 9.60 to the Coalition’s 1.04 at Centrebet from May 9 until May 17, when Labor’s price recovered to 9.00 to 1.05. Sunday night (May 19) saw the Centrebet price move again to 8.50/1.06, and to 8.00/1.07 on Monday afternoon. Between 5pm and 6pm last night the Centrebet prices moved again, to 6.50/1.10.

You can’t help but wonder who is buying Labor at these kinds of prices. True believers? Or savvy investors laying off their Coalition wagers?

A graph of the implied probabilities from Centrebet, Sportsbet and Tom Waterhouse appears below (click for a bigger view):

The Nielsen poll (and media reporting on it) is the likely culprit of Monday’s movement. And yet more evidence that poll results predict betting market predict election outcomes.

Kevin Rudd’s blog post re his evolution on same-sex marriage has a publication date on 7.15pm; if that is a valid timestamp, it suggests that Rudd’s announcement didn’t move the Centrebet price.

Betfair is also showing some modest recovery in the Labor price. The last matched prices there are 10.00/1.11.

Sportsbet and Tom Waterhouse have been exhibiting much less movement in the Federal betting market prices, which you can’t help but interpret as indicative of those bookies handling less volume than Centrebet.

I’d also point out that Centrebet has tended to have a little less profit margin built into its prices than Sportsbet or Waterhouse: for instance, with the prices currently on offer, Centrebet’s overround is about 6.3% vs 7.7% for Sportsbet and Waterhouse. In turn, this might help explain what looks like a slightly more lively market at Centrebet relative to these two competitors.

Removed CRANberries: Package iGenomicViewer (with last version 2.4.8) was removed from CRAN

2013-05-20T17:12:00+00:00

Previous versions (as known to CRANberries) which should be available via the Archive link are:

2012-04-06 2.4.8
2009-11-17 2.4.6
2009-10-29 2.4.4
2009-09-18 2.4.2

Removed CRANberries: Package ccems (with last version 1.03) was removed from CRAN

2013-05-20T15:12:00+00:00

Previous versions (as known to CRANberries) which should be available via the Archive link are:

2009-06-14 1.03
2009-03-21 1.02
2009-03-14 1.01
2009-03-06 1.0

Removed CRANberries: Package odesolve (with last version 0.9-9) was removed from CRAN

2013-05-20T15:12:00+00:00

Previous versions (as known to CRANberries) which should be available via the Archive link are:

2012-03-19 0.9-9
2008-09-16 0.5-20
2008-02-04 0.5-19
2007-10-09 0.5-18
2007-07-18 0.5-7.1
2007-07-18 0.5-17
2006-08-05 0.5-16

Removed CRANberries: Package TreePar (with last version 2.3) was removed from CRAN

2013-05-20T15:12:00+00:00

Previous versions (as known to CRANberries) which should be available via the Archive link are:

2012-08-05 2.3
2011-11-02 2.2
2011-08-23 2.1.2
2011-07-23 2.1
2011-05-10 1.4
2011-03-30 1.3
2011-03-21 1.2
2011-01-25 1.1
2010-06-18 1.0

CRANberries: New package Comp2ROC with initial version 1.0

2013-05-20T09:12:00+00:00

Package: Comp2ROC
Type: Package
Title: Compare two ROC curves that intersect
Version: 1.0
Date: 2012-06-26
Author: Ana C. Braga, with contributions from Hugo Frade
Maintainer: Ana C. Braga
Description: Package that allow you to compare two ROC curves throughout the methodology propose by Ana C. Braga
License: GPL-2
Depends: R (>= 2.15.1), ROCR, boot
Packaged: 2013-05-19 22:04:20 UTC; hugofrade
NeedsCompilation: no
Repository: CRAN
Date/Publication: 2013-05-20 10:36:23

More information about Comp2ROC at CRAN

CRANberries: New package msgpackR with initial version 1.0

2013-05-19T17:12:00+00:00

Package: msgpackR
Type: Package
Title: A library to serialize or unserialize data in MessagePack format
Version: 1.0
Date: 2013-05-19
Author: Mikiya TANIZAWA
Maintainer: Mikiya TANIZAWA 1970mix@gmail.com>
Description: This is the library that can serialize or unserialize MessagePack format data.
License: BSD
Packaged: 2013-05-19 14:42:23 UTC; mikiya
NeedsCompilation: no
Repository: CRAN
Date/Publication: 2013-05-19 18:49:09

More information about msgpackR at CRAN

CRANberries: New package TCC with initial version 1.1.3

2013-05-19T09:12:00+00:00

Package: TCC
Type: Package
Title: TCC: Differential expression analysis for tag count data with robust normalization strategies
Version: 1.1.3
Author: Sun Jianqiang, Tomoaki Nishiyama, Kentaro Shimizu, and Koji Kadota
Maintainer: Tomoaki Nishiyama
Description: This package provides functions for performing differential expression analysis using differentially expressed gene elimination strategy. A simple unified interface is provided which encapsulates functions to calculate normalization factors and estimate differentially expressed genes defined in edgeR, baySeq, and DESeq packages. The appropriate combination provided by TCC allows a more robust and accurate estimation performed easily than directly using original packages. Functions to produce simulation data under various conditions and to plot the data are also provided.
Depends: R (>= 2.15), methods, Biobase(>= 2.13.11), locfit
Imports: ROC, edgeR, baySeq, DESeq
License: GPL-2
Copyright: Authors listed above
URL: http://www.iu.a.u-tokyo.ac.jp/~kadota/TCC
Packaged: 2013-05-19 06:13:44 UTC; tomoaki
NeedsCompilation: no
Repository: CRAN
Date/Publication: 2013-05-19 09:12:46

More information about TCC at CRAN

CRANberries: New package probsvm with initial version 1.00

2013-05-19T07:12:00+00:00

Package: probsvm
Type: Package
Title: probsvm: Class probability estimation for Support Vector Machines
Date: 2013-5-16
Version: 1.00
Author: Chong Zhang, Seung Jun Shin, Junhui Wang, Yichao Wu, Hao Helen Zhang, and Yufeng Liu
Maintainer: Chong Zhang
Depends: kernlab
Description: This package provides multiclass conditional probability estimation for the SVM, which is distributional assumption free.
License: GPL-2
LazyLoad: yes
Packaged: 2013-05-18 23:48:46 UTC; NC
NeedsCompilation: no
Repository: CRAN
Date/Publication: 2013-05-19 08:15:01

More information about probsvm at CRAN

CRANberries: New package pom with initial version 1.0

2013-05-19T07:12:00+00:00

Package: pom
Type: Package
Title: POM - Patch Occupancy Models
Version: 1.0
Date: 2012-11-16
Author: Fawn Hornsby, Ryan Nielson, and Trent McDonald (www.west-inc.com)
Maintainer: Fawn Hornsby
Depends: matrixcalc
Description: This package fits a patch occupancy model
License: GNU General Public License
Packaged: 2013-05-18 20:11:05 UTC; fhornsby
NeedsCompilation: no
Repository: CRAN
Date/Publication: 2013-05-19 08:14:59

More information about pom at CRAN

Removed CRANberries: Package MethComp (with last version 1.15) was removed from CRAN

2013-05-18T13:12:00+00:00

Previous versions (as known to CRANberries) which should be available via the Archive link are:

2012-03-29 1.15
2011-02-18 1.3

CRANberries: New package wpp2008 with initial version 1.0-0

2013-05-18T07:12:00+00:00

Package: wpp2008
Version: 1.0-0
Date: 2013-5-17
Title: World Population Prospects 2008
Author: Hana Sevcikova , Patrick Gerland
Maintainer: Hana Sevcikova
Depends: R (>= 2.14.2)
Description: Data from the United Nation's World Population Prospects 2008
License: GPL (>= 2)
URL: http://esa.un.org/wpp
Packaged: 2013-05-18 00:54:05 UTC; hana
NeedsCompilation: no
Repository: CRAN
Date/Publication: 2013-05-18 08:11:00

More information about wpp2008 at CRAN

CRANberries: New package SMFI5 with initial version 1.0

2013-05-18T01:12:00+00:00

Package: SMFI5
Type: Package
Title: R functions and data from Chapter 5 of 'Statistical Methods for Financial Engineering'
Version: 1.0
Date: 2013-05-16
Author: Bruno Remillard
Maintainer: Bruno Remillard
Description: R functions and data from Chapter 5 of 'Statistical Methods for Financial Engineering', by Bruno Remillard, CRC Press, (2013).
License: GPL (>= 2)
Depends: ggplot2, reshape, corpcor
URL: http://www.r-project.org, http://www.brunoremillard.com
Packaged: 2013-05-17 22:43:07 UTC; 49009427
NeedsCompilation: no
Repository: CRAN
Date/Publication: 2013-05-18 01:56:20

More information about SMFI5 at CRAN

CRANberries: New package CfEstimateQuantiles with initial version 1.0

2013-05-18T01:12:00+00:00

Package: CfEstimateQuantiles
Type: Package
Title: Estimate quantiles using any order Cornish-Fisher expansion
Version: 1.0
Date: 2013-05-12
Author: Maxim Yurchuk
Maintainer: Maxim Yurchuk
Description: Estimate quantiles using formula (18) from http://www.jaschke-net.de/papers/CoFi.pdf (Yaschke; 2001)
License: GPL-2
LazyLoad: yes
Packaged: 2013-05-17 23:54:51 UTC; m2x
NeedsCompilation: no
Repository: CRAN
Date/Publication: 2013-05-18 02:04:44

More information about CfEstimateQuantiles at CRAN

Revolutions: Because it's Friday: Compendium of Logical Fallacies

David Smith — 2013-05-17T20:44:57+00:00

Ahh, arguing on the Internet. Whether it's about politics, religion, culture or science, if you're anything like me you've had 1000 online arguments and changed the mind of exactly nobody. Still, it's fun to exercise those debating muscles, and now with the handy website thou shalt not commit logical fallacies you have the perfect riposte to any invalid, inappropriate or just plain wrong argument (click for the full version):

Someone impugning your character instead of responding to your well-reasoned point? Send 'em over to Ad-Hominem land. Asking you to prove a negative, that they introduced? That's not where the burden of proof lies, buddy. You might say that no true debater would use a tool like this, but I say that's just moving the goalposts.

Each of the fallacies comes with a pithy example. (Begging the Question: "The word of Zorbo the Great is flawless and perfect. We know this because it says so in The Great and Infallible Book of Zorbo's Best and Most Truest Things that are Definitely True and Should Not Ever Be Questioned."). You can find all of the fallacies and examples by downloading the poster, which comes under a sharing-friendly Creative Commons license. You can also explore each of the fallacies at yourlogicalfallacyis.com.

Have a great weekend!

Revolutions: R 3.0.1 released

David Smith — 2013-05-17T18:40:38+00:00

The R core group has quickly followed up with a patch to R version 3. Announced yesterday, R 3.0.1 (code name: "Good Sport") improves serialization performance with big objects, improves reliability for parallel programming and fixes a few minor bugs. (You can find the complete list of changes in the NEWS file.) The source distribution and Windows and Linux binaries are available for download now from your local mirror, and the Mac binary will follow soon.

The R core group continues to develop R version 3, and more patches can be expected over the coming year. Revolution R Enterprise will update to inclde R version 3 in late 2013.

R-announce: R 3.0.1 is released

Removed CRANberries: Package nlreg (with last version 1.2-1) was removed from CRAN

2013-05-17T17:12:00+00:00

Previous versions (as known to CRANberries) which should be available via the Archive link are:

2011-11-25 1.2-1
2009-10-05 1.2-0
2009-09-24 1.1-3

Removed CRANberries: Package cond (with last version 1.2-1) was removed from CRAN

2013-05-17T17:12:00+00:00

Previous versions (as known to CRANberries) which should be available via the Archive link are:

2011-11-25 1.2-1
2009-10-05 1.2-0
2009-09-24 1.1-3

Removed CRANberries: Package BAYSTAR (with last version 0.2-5) was removed from CRAN

2013-05-17T17:12:00+00:00

Previous versions (as known to CRANberries) which should be available via the Archive link are:

2012-03-20 0.2-5
2009-10-09 0.2-3
2009-09-24 0.2-2
2008-08-08 0.1-2
2008-01-28 0.1-1
2008-01-07 0.1-0

Removed CRANberries: Package rv (with last version 2.2.1) was removed from CRAN

2013-05-17T17:12:00+00:00

Previous versions (as known to CRANberries) which should be available via the Archive link are:

2013-01-15 2.2.1
2012-01-27 2.0.0
2011-02-01 1.1.0
2008-08-11 1.0
2007-10-10 0.949
2007-08-25 0.946
2007-07-10 0.945
2006-10-27 0.925

Removed CRANberries: Package csampling (with last version 1.2-1) was removed from CRAN

2013-05-17T17:12:00+00:00

Previous versions (as known to CRANberries) which should be available via the Archive link are:

2011-11-25 1.2-1
2009-10-05 1.2-0
2009-09-24 1.1-3

Removed CRANberries: Package NADA (with last version 1.5-4) was removed from CRAN

2013-05-17T17:12:00+00:00

Previous versions (as known to CRANberries) which should be available via the Archive link are:

2012-03-16 1.5-4
2010-12-22 1.5-3
2009-05-24 1.5-2
2007-09-27 1.5-1
2006-12-19 1.5-0

Removed CRANberries: Package diff (with last version 1.0.2) was removed from CRAN

2013-05-17T17:12:00+00:00

Previous versions (as known to CRANberries) which should be available via the Archive link are:

2011-01-25 1.0.2
2011-01-24 1.0.1
2011-01-15 1.0