Macroeconomists using large datasets often face the choice of working with either a vector autoregression (VAR) or factor model. In this paper, we develop conjugate Bayesian VAR subspace shrinkage prior that combines two. This shrinks towards which is defined by Our approach allows for estimating strength and number factors. After establishing theoretical properties our prior, show it successfu...

Estimation and structure learning of high-dimensional signals via a normal sequence model are considered, where the underlying parameter vector is piecewise constant, or has block structure. A Bayesian fusion estimation method developed by using Horseshoe prior to induce strong shrinkage effect on successive differences in mean parameters, simultaneously imposing sufficient concentration for no...

To improve feature learning ability and accurately diagnose the faults of rolling bearings under a strong background noise environment, we present new shrinkage function named leaky thresholding to replace soft in deep residual networks (DRSNs). In this work, discover that such improved (IDRSNs) can be realized by using group searching method optimize slope value thresholding, IDRSNs more effec...

Thresholding estimators in an orthonormal wavelet basis are well established tools for Gaussian noise removal. However, the threshold choice ̧ = 3⁄4 p (2logN) suggested by Donoho and Johnstone, implies the knowledge of the variance 3⁄4 of the noise. In this talk we consider the denoising problem as a variational problem, whose solution can be formulated in terms of wavelet shrinkage, and we prop...

The shrinkage estimate of a quantitative trait locus (QTL) effect is the posterior mean of the QTL effect when a normal prior distribution is assigned to the QTL. This note gives the derivation of the shrinkage estimate under the multivariate linear model. An important lemma regarding the posterior mean of a normal likelihood combined with a normal prior is introduced. The lemma is then used to...

We consider Bayesian nonparametric estimation of a survival time subject to right-censoring in the presence potentially high-dimensional predictors. argue that several approaches, such as random forests and existing possess drawbacks, including: computational difficulties; lack known theoretical properties; ineffectiveness at filtering out irrelevant propose two models based on additive regress...