I needed to look up a formula for the estimating the relative efficiency to a completely randomized block design to a design without blocking. Continue reading
Monthly Archives: March 2014
PMean: Using statistical design principles to plan a Monte Carlo analysis
I want to run a Monte Carlo analysis of various Bayesian estimators to see how they perform when the prior distribution is “wrong”. I’m like everyone else–I just plunge in and start. But halfway through the Monte Carlo analysis, I realized that I could make my life easier and produce a better quality Monte Carlo analysis if I used basic statistical design principles. Here’s a brief outline of some of these design principles. Continue reading
PMean: Comparing two vectors with possible missing values
I want to compare two vectors in R that have the same length and identify where, if anywhere, that the two vectors don’t match. It sounds like an easy thing to do, but missing values muddy things up quite a bit. Continue reading
Recommended: foreach: Foreach looping construct for R
This is an add-on package for R that allows you to separate a lengthy programming task into different pieces, each of which can run on a separate core of your multi-core computer. Continue reading
Recommended: Meet Charles J. Rothwell, Director, National Center for Health Statistics
This is a brief interview with the director of the National Center for Health Statistics, a statistical agency for the United States government that produces many nationally representative health surveys. Continue reading
Recommended: Qualifications for Teaching an Introductory Statistics Course
The American Statistical Association and the Mathematical Association of America published a joint statement on the qualifications that a college teacher would need in order to teach an introductory statistics course for undergraduate students. This is an issue in many Mathematics Departments which might need to teach such a class but would not be big enough to have a degreed Statistician. A minimum amount of course work is needed, but also practical experience with data analysis, which might come from “advanced courses, projects, consulting, or research.” Continue reading