Modeling With Data

Today, I am a blog.

PDF version

Hi.

This is the blog for my book, Modeling with Data. That means that I'll be discussing statistics, scientific inquiry, computing, academia, publishing, and what I had for lunch before sitting down to write.

Let me tell you why I'm setting up another blog: because statistics is amazing. Seriously enthralling. Pure mathematics is internally consistent, but comfortably ignores what we like to call reality; statistics is a field that does the dirty work of linking mathematical models with observed phenomena. This is a hard problem, not in the sense that with enough elbow grease we can solve it, but in the sense that we mortals are fundamentally incapable of a definitive solution.

Of course, we do our best anyway. What are the odds that it will rain tomorrow? Your weatherman will be happy to give you a number, even though just explaining what that question means is about impossible.

Modeling with Data grazed the surface of these questions, but it's a textbook, so I focused on specific techniques and details rather than the social, technological, and even philosophical issues underlying the techniques. I will not be talking about the politics of information, because I already have a blog about that. Also, despite being the author of a statistics textbook, I'm human, and can perhaps share with you some interesting ideas about writing, publishing, and doing research in the present day.

Finally, an important piece of logistics: I don't write 140-character posts every fifteen minutes, but instead tend toward infrequent, more substantive posts. So don't click back here every day to see if something is new: get notifications of new posts via the RSS feed. If you don't know what RSS is, it's the civilized way to read everything from The New York Times to The Onion.

[]
[Next entry: "Moore's law won't save you"]