No Free Lunch
The average is the modern solution to summarising data. Where more than one measurement existed, then the original historical reflex was to choose the best measurement: a fine strategy if you know which is best. Originally, God, the Church, or the King (as His official representative) would choose. Practically, this meant the ‘most expensive’ measurement (see Stephen Siegler).
Simpson's Paradox backwards
I have been enjoying reading Judea Pearl’s new book Causal Inference in Statistics: A primer1
Pearl J. Causal Inference in Statistics. John Wiley & Sons; 2016 ↩
R functions you should know
A running list of R functions that I should have known, and didn’t with a minimal worked example in the style of tl;dr.
Parsing command line options in R
Notes to myself in the form of an R script about how to parse command line arguments. See StackOverflow for alternatives.
Get up, git up
I have been using SourceTree as a GUI for git, but just came across GitUp. It starts off just looking like a pretty way to view your repository with a fairly typical graph, but you can in fact work from within the graph.
A curated list of awesome R frameworks, packages and software.
Globally set options in R markdown
Normally you specify options in R markdown at the start of each chunk which requires typing them out for every block if they differ from the defaults.
Getting data into R
Science isn't broken
The important lesson here is that a single analysis is not sufficient to find a definitive answer. Every result is a temporary truth, one that’s subject to change when someone else comes along to build, test and analyze anew.
A day in the life ... visualised