R

Since starting to use R about a decade ago, I have been a consistent consumer of information from stackoverflow (SO). It is an invaluable resource and I am grateful to those who take the time to ask good questions and provide good answers. I have also felt a little guilty for not giving back by being an active participant. In this post I will give some (mostly bad) reasons why I was not actively participating, what finally pushed me to participate, and some benefits and tips for answering questions on SO.

Conditioning on robust summaries of the data in a Bayesian model is one way to achieve robustness to model miss-specification. I have called this the “restricted likelihood” here and here since the full data likelihood is replaced with the likelihood conditioned on only the robust summary (i.e. a restricted likelihood). One of the easiest examples to conceptualize is outliers in a univariate setting. Suppose the true data generating mechanism is a contaminated normal:

The tibble package within the the tidyverse provides ‘a modern take on data frames.’ It’s loaded with nice features, one of which is the ability to store list-columns. List-columns provide a concise way to store lists within a row of a data frame. In particular, this is useful for storing functional data because a common feature in such data is that each function is not collected at the same number of points.