http://www.arnoldkling.com/blog/hal-varian-on-big-data/ taking the most charitable view of those who disagree Mon, 21 Dec 2020 14:00:34 +0000 hourly 1 https://wordpress.org/?v=4.0.32 http://www.arnoldkling.com/blog/hal-varian-on-big-data/#comment-306018 Tue, 17 Dec 2013 20:25:36 +0000 http://www.arnoldkling.com/blog/?p=2389#comment-306018 Nonsense jargon aside (support vector machines??), the different methods are a fulfilling what Ed Leamer wrote about 35 years ago. You have to find a way to, first, compute all possible regressions given the data and plausible theory, and then present all possible regression results in a way to establish whether or not there is a robust and meaningful finding. This is “avoid cherry-picking” taken to its logical conclusion.

]]>