The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Online Program Home
Abstract Details
|
Activity Number:
|
481
|
|
Type:
|
Invited
|
|
Date/Time:
|
Wednesday, August 1, 2012 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
WNAR
|
| Abstract - #303525 |
|
Title:
|
Forest Density Estimation
|
|
Author(s):
|
Han Liu*+
|
|
Companies:
|
The Johns Hopkins University
|
|
Address:
|
615 N Wolfe St, Baltimore, MD, 21205, United States
|
|
Keywords:
|
nonparametric inference ;
high dimensions ;
graphical models ;
unsupervised learning ;
data visualization ;
Forest density estimation
|
|
Abstract:
|
We study graph estimation and density estimation in high dimensions, using a family of density estimators based on forest structured undirected graphical models. For density estimation, we do not assume the true distribution corresponds to a forest; rather, we form kernel density estimates of the bivariate and univariate marginals, and apply Kruskal's algorithm to estimate the optimal forest. In terms of function estimation, this method achieves the minimax optimal rate of convergence. In terms of graph estimation, this method even achieves the optimal parametric rate of convergence. Therefore, the extra flexibility gained by nonparametric modeling comes at very low cost. Different variants and connections of the forest models are proposed. The performance of these methods is illustrated and compared on several real and simulated examples.
|
The address information is for the authors that have a + after their name.
Authors who are presenting talks have a * after their name.
Back to the full JSM 2012 program
|
2012 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Continuing Education program, please contact the Education Department.