• Corpus ID: 235490088

Scalable Bayesian change point detection with spike and slab priors

  title={Scalable Bayesian change point detection with spike and slab priors},
  author={Lorenzo Cappello and Oscar Hernan Madrid Padilla and Julia A. Palacios},
  journal={arXiv: Methodology},
We study the use of spike and slab priors for consistent estimation of the number of change points and their locations. Leveraging recent results in the variable selection literature, we show that an estimator based on spike and slab priors achieves optimal localization rate in the multiple offline change point detection problem. Based on this estimator, we propose a Bayesian change point detection method, which is one of the fastest Bayesian methodologies, and it is more robust to… 

Figures and Tables from this paper

Controlled Discovery and Localization of Signals via Bayesian Linear Programming

Bayesian Linear Programming (BLiP) is proposed, a method which overcomes this intractability to jointly detect and localize signals in a way that nearly maximizes the expected resolution-adjusted power while provably controlling false positives.

Variance estimation in graphs with the fused lasso

The problem of variance estimation in general graph-structured problems is studied and a linear time estimator for the homoscedastic case is developed that can consistently estimate the variance in general graphs and a simple total variation regularization estimator is studied for estimating the signal of variances in the heteroscedastics case.

Quantile Regression by Dyadic CART

The scope of these globally optimal regression tree based methodologies to be applicable for heavy tailed data is extended and the QDCART estimator enjoys adaptively rate optimal estimation guarantees for piecewise constant and bounded variation function classes.

Changepoint Detection: An Analysis of the Central England Temperature Series

: This paper presents a statistical analysis of structural changes in the Central England temperature series, one of the longest surface temperature records available. A changepoint analysis is



Bayesian variable selection with shrinking and diffusing priors

We consider a Bayesian approach to variable selection in the presence of high dimensional covariates based on a hierarchical model that places prior distributions on the regression coefficients as

Exact posterior distributions and model selection criteria for multiple change-point detection problems

This work derives exact, explicit and tractable formulae for the posterior distribution of variables such as the number of change-points or their positions and demonstrates that several classical Bayesian model selection criteria can be computed exactly.

Heterogeneous change point inference

We propose, a heterogeneous simultaneous multiscale change point estimator called ‘H‐SMUCE’ for the detection of multiple change points of the signal in a heterogeneous Gaussian regression model. A

Stepwise Signal Extraction via Marginal Likelihood

A maximum marginal likelihood estimator is formulated, which can be computed with a quadratic cost using dynamic programming and is applicable to a wide range of models and offers appealing results in practice.

Optimal nonparametric change point detection and localization.

It is proved that the procedure based on wild binary segmentation is nearly minimax rate-optimal and a phase transition in the space of model parameters that separates parameter combinations for which consistent localization is possible from the ones for which this task is statistical unfeasible is demonstrated.

Nonparametric maximum likelihood approach to multiple change-point problems

In multiple change-point problems, different data segments often follow different distributions, for which the changes may occur in the mean, scale or the entire distribution from one segment to

Change-point detection in high-dimensional covariance structure

A novel approach to detection of an abrupt change in the covariance structure of a high-dimensional random process, which has applications in many areas e.g., neuroimaging and finance, is introduced and theoretically justified under mild assumptions.

The horseshoe estimator for sparse signals

This paper proposes a new approach to sparsity, called the horseshoe estimator, which arises from a prior based on multivariate-normal scale mixtures. We describe the estimator's advantages over

Multiscale change point inference

A new estimator, the simultaneous multiscale change point estimator SMUCE, is introduced, which achieves the optimal detection rate of vanishing signals as n→∞, even for an unbounded number of change points.

Locally Adaptive Smoothing with Markov Random Fields and Shrinkage Priors.

This method uses shrinkage priors to induce sparsity in order-k differences in the latent trend function, providing a combination of local adaptation and global control, and finds the horseshoe prior provides the best compromise between bias and precision.