I frequently use lattice and ggplot2 to create trellis/faceted graphics. But, I gave up using these packages in a recent application, where I had initially constructed a complex graphic using the base R plotting functions. When I later decided that I wanted a faceted version, there was a dilema: re-create the complex graphic using lattice [...]
There was an R-help thread in January regarding text in the righthand margin of an R graphic, where the text should be rendered in reading order from top to bottom. The base R function mtext is used to plot text in the margin. But, mtext is only able to render text from left to right [...]
A new Annals of Internal Medicine article describes a study that compares two employer-sponsored financial incentive programs for promoting weight loss among obese employees. I first read about the article at the Pacific Standard. The study design is a randomized controlled prospective trial. The two programs are as follows: Program 1. Obese employees are given [...]
JW Emerson, WA Green, B Schloerke, J Crowley, D Cook, H Hofmann, H Wickham (2013) The Generalized Pairs Plot. Journal of Computational and Graphical Statistics 22(1). Here's a free preprint version. Until this new paper and implementation by Emerson et al., there were no widely available pairs plots that accommodated both numerical and categorical fields. [...]
I cringe when I see research proposals that describe a sophisticated statistical approach, yet do not evaluate this approach in their power/precision/sample size planning. It's often the case that a simplified version of the proposed statistical approach is used instead. Presumably, this is due to the limited availability of power/precision/sample size planning software for sophisticated [...]
Due largely to the work of Clint Cummins, the sas7bdat file format has become a bit less shrouded. In particular, we now know the following: how to detect files with compressed data (and fail graciously) more details about the platform that generated the file (e.g., endianess, OS details) how to read files that were generated [...]
I have sometimes heard complaints from collaborators that it will be impossible to have their work published in the mainstream literature unless a p-value is reported. This post is to report yet another counterexample that was recently published; a meta-analysis for the odds of perioperative bleeding complications in patients taking one of several anticoagulant/antiplatelet drugs. [...]
Some time ago, I had an ineloquent and less-than-cordial online discussion with a commenter on this site, partially about how statisticians define the term "parameter". This post is just to quote a relevant passage from "Bootstrap Methods and Their Application", by Davison and Hinkley (1997), that better articulates a point I had made earlier. 2.1.1 [...]
That's a mouthful! I presented this topic to a group of Vandy statisticians a few days ago. My notes (essentially reproduced in this post) are recorded at the Dept. of Biostatistics wiki: HowToBootstrapCorrelatedData. The presentation covers some bootstrap strategies for hierarchically structured (correlated) data, but focuses on the multi-stage bootstrap; an extension of that described [...]
A central theme of Don Fraser's article, titled "Is Bayes Posterior just Quick and Dirty Confidence?", was that Bayesian confidence regions have approximate, and sometimes poor frequentist coverage (i.e., the frequency with which a confidence region contains the true parameter value under repeated sampling). Fraser has this warning: The failure to make true assertions with [...]
