"Significance Tests for Adaptive Modelling" (Today at the Statistics Seminar)
Attention conservation notice: Late notice of a very
technical presentation about theoretical statistics in a city you don't live in.
Today's speaker needs no introduction for those interested
in modern, high-dimensional
statistics (but will get an introduction anyway):
- Robert Tibshirani,
"Significance Tests for Adaptive Modelling"
- Abstract: In this talk we consider testing the significance of the
terms in a fitted regression, fit via the
lasso. We propose a novel test
statistic for this problem, and show that it has a simple asymptotic null
distribution. This work builds on the least angle regression approach for
fitting the lasso, and the notion of degrees of freedom for adaptive models
(Efron 1986) and for the lasso (Efron et al. 2004; Zou et al. 2007). We give
examples of this procedure and discuss extensions to generalized linear models.
In the second part of the talk we will discuss
extensions of this work to graphical
models, specifically inference for connected components.
- Joint work with Richard
Taylor, Ryan Tibshirani,
and Max Jacob Grazier G'Sell.
- Time and place: 4:30--5:30 pm on Monday, 21 October 2013, in Baker Hall 136A
As always, the talk is free and open to the public.
Enigmas of Chance
Posted at October 21, 2013 13:55 | permanent link