You’ll find links to Cathy O’Neil’s important book in the Blogroll here, as well as a link to reviews of it.
I have not read it yet. While I have pre-ordered it, it’s not available. I have read the reviews, all favorable. I suspect it expresses a lot of concern that statisticians, including myself, have long expressed regarding “data mining” (data dredging), and inferences derived from “big data”, including a lack of concern about the quality of the data, forgetting about cleaning, and whether or not, despite the size of a dataset, there is enough data to warrant a strong estimate or inference in a particular case.
These arguments have been made by many, like Simply Statistics‘ “Why big data is in trouble: they forgot about applied statistics” from 2014. Or Bajorski’s “Applied Statistics Comes to the Rescue of Big Data”. And the wave of emphasis upon pure machine learning methods of late does not help: These obfuscate and confound, making it harder to understand what’s really going on.
Anyway, I very much look forward to reading the book!
Yo, everyone! The final version of my book now exists, and I have exactly one copy! Here’s my editor, Amanda Cook, holding it yesterday when we met for beers:
Here’s my son holding it:
He’s offered to become a meme in support of book sales.
Here’s the back of the book, with blurbs from really exceptional people:
In other exciting book news, there’s a review by Richard Beales from Reuter’s BreakingViews, and it made a list of new releases in Scientific American as well.
I want to apologize in advance for all the book news I’m going to be blogging, tweeting, and otherwise blabbing about. To be clear, I’ve been told it’s my job for the next few months to be a PR person for my book, so I guess that’s what I’m up to. If you come here for ideas and are turned off by cheerleading, feel…
View original post 34 more words