Thoughts on Statistics and Machine Learning
In addition to being the best comedy TV show ever, Seinfeld was a great source of wisdom. In one episode, Jerry counsels George: When you hit that high note, you say goodnight and walk off. Later, George has a good line at a meeting at Kruger Industrial Smoothing. Then he says: Alright! That s it for [ ]
I have mentioned Stein s method in passing, a few times on this blog. Today I want to talk about Stein s method in a bit of detail. 1. What Is Stein s Method? Stein s method, due to Charles Stein, is actually quite old, going back to 1972. But there has been a great deal of interest in [ ]
On Monday we had an interesting seminar by Samory Kpotufe on nonparametric regression. Samory presented a method for choosing the smoothing parameter, locally, in nonparametric regression. The method is simple and intuitive: construct confidence intervals using many different values of the smoothing parameter. Choose the value at which the confidence intervals stop intersecting. He has [ ]
Don t you wish there was a way to comment on papers? Now there is. Thanks to the efforts of Rob Tibshirani, Pat Brown, Mike Eisen, David Lipman and others there is now a system called PubMed Commons. PubMed is the central repository for biomedical research. PubMed Commons allows people to have active discussions of papers. [ ]
I am visiting the Department of Statistics at the University of Washington today and am enjoying their warm hospitality (thanks Emily!). I have a little breathing room now so I thought I would write a quick blog post. Earlier this year, I posted my contribution to the collection celebrating the 50th anniversary of The Committee [ ]
Richard Lockhart, Jonathan Taylor, Ryan Tibshirani and Rob Tibshirani have an interesting paper about significance tests for the lasso. The paper will appear in The Annals of Statistics. I was asked to write a discussion about the paper. Here is my discussion. (I suggest your read their paper before reading my discussion.) Assumption-Free High-Dimensional Inference: [ ]
Mladen Kolar, Alessandro Rinaldo and I have uploaded a paper to arXiv entitled Estimating Undirected Graphs Under Weak Assumptions. As the name implies, the goal is to estimate an undirected graph from random vectors . Here, each is a vector with coordinates, or features. The graph has nodes, one for each feature. We put an [ ]
A couple of announcements: First: A message from Jeff Leek: We are hosting an unconference on Google Hangouts. We got some really amazing speakers to talk about the future of statistics. I wonder if you could help advertise the unconference on your blogs. Here is our post:…
There are many ways to discuss the quality of estimators in statistics. Today I want to review three common notions: presistency, consistency and sparsistency. I will discuss them in the context of linear regression. (Yes, that s presistency, not persistency.) Suppose the data are where , and . Let be an estimator of . Probably the [ ]
Time for a provocative post. There is a nice YouTube video with Tony O Hagan interviewing Dennis Lindley. Of course, Dennis is a legend and his impact on the field of statistics is huge. At one point, Tony points out that some people liken Bayesian inference to a religion. Dennis claims this is false. Bayesian inference, [ ]