March 5, 2015
Briefly
- “Yesterday the Herald reported on some confusion about what’s happening to New Zealand property prices. QV said prices were continuing their seemingly inexorable rise, while Barfoot & Thompson said that in Auckland prices had dipped, based on its own sales.” Aaron Schiff, arguing that housing price data should be open
- Another tea story, but this one does say what experiments were done and what the equivalent dose would be in humans.
- Predictive models in language: autocompleting to ‘Qatar’
- Predictive models in language: detecting ‘fake’ (or Gaelic) names on Facebook
- Predictive vs causal explanations in language, from XKCD
Thomas Lumley (@tslumley) is Professor of Biostatistics at the University of Auckland. His research interests include semiparametric models, survey sampling, statistical computing, foundations of statistics, and whatever methodological problems his medical collaborators come up with. He also blogs at Biased and Inefficient See all posts by Thomas Lumley »
I’m a great believer in open data but feel the need to point out that “free” (open) data is not “free” (costless). It’s like saying roads should be free to use, which of course means that some users do better than others out of it.
I also point out a recent RatSWD working paper #246, “A Reputation Economy: Results from an Empirical Survey on Academic Data Sharing” on the views of academics sharing their data – the cost of doing so being one hold up.
10 years ago