Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
5129637 | Statistics & Probability Letters | 2017 | 7 Pages |
Abstract
We study the piece-wise smooth regression from a theoretical Bayesian perspective. Our results indicate that under some mild assumptions, the posterior of the regression model and the change-points locations contracts at optimal nonparametric convergence rate up to a log-factor, and the number of change-points is posterior consistent.
Related Topics
Physical Sciences and Engineering
Mathematics
Statistics and Probability
Authors
Taihe Yi, Zhengming Wang,