Bayesian shrinkage methods for partially observed data with many predictors

DOI10.1214/13-AOAS668MaRDI QIDQ2441860zbMATH OpenWikidataFDO

Authors Philip S. Boonstra, Bhramar Mukherjee, Jeremy M. G. Taylor

Publication date 28 March 2014

Published in The Annals of Applied Statistics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1401.2324, https://projecteuclid.org/euclid.aoas/1387823319

zbMATH Keywords

Markov chain Monte Carlohigh-dimensional datamissing datameasurement errors

Mathematics Subject Classification ID

Bayesian inference (62F15) Linear regression; mixed models (62J05) Ridge regression; shrinkage estimators (Lasso) (62J07) Biochemistry, molecular biology (92C40) Empirical decision procedures; empirical Bayes procedures (62C12)

Abstract: Motivated by the increasing use of and rapid changes in array technologies, we consider the prediction problem of fitting a linear regression relating a continuous outcome $Y$ to a large number of covariates $mathbfX$, for example, measurements from current, state-of-the-art technology. For most of the samples, only the outcome $Y$ and surrogate covariates, $mathbfW$, are available. These surrogates may be data from prior studies using older technologies. Owing to the dimension of the problem and the large fraction of missing information, a critical issue is appropriate shrinkage of model parameters for an optimal bias-variance trade-off. We discuss a variety of fully Bayesian and Empirical Bayes algorithms which account for uncertainty in the missing data and adaptively shrink parameter estimates for superior prediction. These methods are evaluated via a comprehensive simulation study. In addition, we apply our methods to a lung cancer data set, predicting survival time ($Y$) using qRT-PCR ($mathbfX$) and microarray ($mathbfW$) measurements.

Recommendations

• Empirical Bayes estimates for large-scale prediction problems
• A pseudo-Bayesian shrinkage approach to regression with missing covariates
• Univariate shrinkage in the Cox model for high dimensional data
• Bayesian methods for regression using surrogate variables
• Bayesian Semiparametric Multiple Shrinkage

Cites work

• scientific article; zbMATH DE number 4159863 (Why is no real title available?)
• scientific article; zbMATH DE number 3567782 (Why is no real title available?)
• scientific article; zbMATH DE number 1834445 (Why is no real title available?)
• A Statistical View of Some Chemometrics Regression Tools
• Bayesian shrinkage methods for partially observed data with many predictors
• Covariance-regularized regression and classification for high dimensional problems
• Empirical Bayes Gibbs sampling
• Ridge Regression: Biased Estimation for Nonorthogonal Problems
• Sampling-Based Approaches to Calculating Marginal Densities
• Smoothing noisy data with spline functions: Estimating the correct degree of smoothing by the method of generalized cross-validation
• Stochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of Images
• The Bayesian Lasso
• The Calculation of Posterior Distributions by Data Augmentation

Cited in

(8)

• Bayesian prediction based on a class of shrinkage priors for location-scale models
• Data enriched linear regression
• Bayesian variable selection for multioutcome models through shared shrinkage
• Singular value shrinkage priors for Bayesian prediction
• Bayesian shrinkage methods for partially observed data with many predictors
• On the estimation of homogeneous population size from a complex dual-record system
• A Bayesian precision medicine framework for calibrating individualized therapeutic indices in cancer
• Bayesian Semiparametric Multiple Shrinkage

Describes a project that uses

Uses Software

• scout
• Annotate

This page was built for publication: Bayesian shrinkage methods for partially observed data with many predictors

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2441860)