scholarly journals Bayesian Sparse Spiked Covariance Model with a Continuous Matrix Shrinkage Prior

2022 ◽  
Vol -1 (-1) ◽  
Author(s):  
Fangzheng Xie ◽  
Joshua Cape ◽  
Carey E. Priebe ◽  
Yanxun Xu
Keyword(s):  
Covariance Model ◽  
Matrix Shrinkage ◽  
Shrinkage Prior

scholarly journals Discussion on Competition for Spatial Statistics for Large Datasets

2021 ◽  
Author(s):  
Roman Flury ◽  
Reinhard Furrer
Keyword(s):  
Spatial Statistics ◽  
Covariance Function ◽  
Likelihood Function ◽  
Large Datasets ◽  
Missing Observations ◽  
Covariance Model ◽  
Compactly Supported ◽  
Covariance Tapering ◽  
Simple Kriging ◽  
Estimated Parameters

AbstractWe discuss the experiences and results of the AppStatUZH team’s participation in the comprehensive and unbiased comparison of different spatial approximations conducted in the Competition for Spatial Statistics for Large Datasets. In each of the different sub-competitions, we estimated parameters of the covariance model based on a likelihood function and predicted missing observations with simple kriging. We approximated the covariance model either with covariance tapering or a compactly supported Wendland covariance function.

scholarly journals Bayesian cumulative shrinkage for infinite factorizations

Biometrika ◽  
2020 ◽  
Vol 107 (3) ◽  
pp. 745-752 ◽  
Author(s):  
Sirio Legramanti ◽  
Daniele Durante ◽  
David B Dunson
Keyword(s):  
Monte Carlo ◽  
Factor Analysis ◽  
Monte Carlo Algorithm ◽  
Model Complexity ◽  
Latent Factors ◽  
Shrinkage Process ◽  
Shrinkage Prior ◽  
Performance Gains ◽  
Shrinkage Priors ◽  
Model Dimension

Summary The dimension of the parameter space is typically unknown in a variety of models that rely on factorizations. For example, in factor analysis the number of latent factors is not known and has to be inferred from the data. Although classical shrinkage priors are useful in such contexts, increasing shrinkage priors can provide a more effective approach that progressively penalizes expansions with growing complexity. In this article we propose a novel increasing shrinkage prior, called the cumulative shrinkage process, for the parameters that control the dimension in overcomplete formulations. Our construction has broad applicability and is based on an interpretable sequence of spike-and-slab distributions which assign increasing mass to the spike as the model complexity grows. Using factor analysis as an illustrative example, we show that this formulation has theoretical and practical advantages relative to current competitors, including an improved ability to recover the model dimension. An adaptive Markov chain Monte Carlo algorithm is proposed, and the performance gains are outlined in simulations and in an application to personality data.

Spatiotemporal EEG/MEG source analysis based on a parametric noise covariance model

2002 ◽  
Vol 49 (6) ◽  
pp. 533-539 ◽  
Author(s):  
H.M. Huizenga ◽  
J.C. de Munck ◽  
L.J. Waldorp ◽  
R.P.P.P. Grasman
Keyword(s):  
Source Analysis ◽  
Covariance Model ◽  
Noise Covariance ◽  
Parametric Noise

The Effect of Matrix Shrinkage on Damage Accumulation in Composites

1989 ◽  
Vol 23 (6) ◽  
pp. 570-586 ◽  
Author(s):  
L.S. Penn ◽  
R.C.T. Chou ◽  
A.S.D. Wang ◽  
W.K. Binienda
Keyword(s):  
Damage Accumulation ◽  
Matrix Shrinkage

A supplement on CLT for LSS under a large dimensional generalized spiked covariance model

2018 ◽  
Vol 138 ◽  
pp. 57-65 ◽  
Author(s):  
Jiaqi Chen ◽  
Yangchun Zhang ◽  
Weiming Li ◽  
Boping Tian
Keyword(s):  
Covariance Model
Sign in / Sign up

Export Citation Format

Share Document