Seriously, NWU comparable to Yale and Princeton, even without mentioning MEDS in Kellogg
Northwestern 2018 Job market candidates

Ok, let me do it one more time.
(Bahadur and Savage, 1956) tell us that it is not possible to have a test for the null that the sample mean equals to zero uniformly over a large class of distribution. We need to restrict the class of distributions. If we have this impossibility result for testing the sample mean. How can we expect that more complicated tests will have better properties? I don't understand why we need 100 papers to show that testing something uniformly over a large class of distributions is not possible. (Bahadur and Savage, 1956) was impressive because they were the first to show this impossibility result for a very simple testing problem. But what do we learn from all the subsequent literature?
There exists a trivial test nonparametric test for continuity in the conditional CDF. The permutation test looks less restrictive, because it does not require existence of density everywhere, except locally to the cutoff. Is this case empirically relevant? I also can't take seriously testing papers that do not discuss power.

And this advances Economics in what direction?
Ok, let me do it one more time.
(Bahadur and Savage, 1956) tell us that it is not possible to have a test for the null that the sample mean equals to zero uniformly over a large class of distribution. We need to restrict the class of distributions. If we have this impossibility result for testing the sample mean. How can we expect that more complicated tests will have better properties? I don't understand why we need 100 papers to show that testing something uniformly over a large class of distributions is not possible. (Bahadur and Savage, 1956) was impressive because they were the first to show this impossibility result for a very simple testing problem. But what do we learn from all the subsequent literature?
There exists a trivial test nonparametric test for continuity in the conditional CDF. The permutation test looks less restrictive, because it does not require existence of density everywhere, except locally to the cutoff. Is this case empirically relevant? I also can't take seriously testing papers that do not discuss power. 
Economics is not about economics anymore. It's about statistical tests about fetuses. People choosing on the margin? What, are you stuck in 1955?
That's the point. In what direction his the ET and RESTUD publications advance economics?
And this advances Economics in what direction

These are also very good. In the same league as Daniel.
http://stanford.edu/~shota2/
http://web.stanford.edu/~ichaves/
https://scholar.harvard.edu/xiaoshengmu/home
http://www.jonlib.com/The only question mark is how can so many theorists survive at the same time