Worthwhile reads from the past week:
Show me the prediction interval!
Two very good posts:
Name some problems where frequentists have gotten better results than Bayesians.
Probabilities generate frequencies not the other way round.
– Fernando
That’s not so much a Thought for the Day as it is one of life’s guiding principles.
WWJTD?
(“What would John Tukey do?”)
I’m attending the Computer Vision and Pattern Recognition (CVPR) conference this week. I’ve read a lot of papers from past proceedings but this is the first time I’ve attended. (A few interesting talks and posters but so far nothing particularly applicable to work. Still a few days to go though.) Anyhow, best quote so far from the meeting:
You might think that you can move in any direction in the tomato space.
Ah, but you can’t! In all seriousness, the speaker was talking about a mathematical representation of tomatoes from unripe to ripe to sliced or diced. More generally, his topic was representation of objects undergoing transformations. It was an interesting talk.
“We often forget how science and engineering function. Ideas come from previous exploration more often than from lightning strokes. Important questions can demand the most careful planning for confirmatory analysis. Broad general inquiries are also important. Finding the question is often more important than finding the answer. Exploratory data analysis is an attitude, a flexibility, and a reliance on display, NOT a bundle of techniques, and should be so taught. Confirmatory data analysis, by contrast, is easier to teach and easier to computerize. We need to teach both; to think about science and engineering more broadly; to be prepared to randomize and avoid multiplicity.”
Source: xkcd
