Access

You are not currently logged in.

Access your personal account or get JSTOR access through your library or other institution:

login

Log in to your personal account or through your institution.

Nonconjugate Bayesian Estimation of Covariance Matrices and Its Use in Hierarchical Models

Michael J. Daniels and Robert E. Kass
Journal of the American Statistical Association
Vol. 94, No. 448 (Dec., 1999), pp. 1254-1263
DOI: 10.2307/2669939
Stable URL: http://www.jstor.org/stable/2669939
Page Count: 10
  • Download ($14.00)
  • Cite this Item
Nonconjugate Bayesian Estimation of Covariance Matrices and Its Use in Hierarchical Models
Preview not available

Abstract

The problem of estimating a covariance matrix in small samples has been considered by several authors following early work by Stein. This problem can be especially important in hierarchical models where the standard errors of fixed and random effects depend on estimation of the covariance matrix of the distribution of the random effects. We propose a set of hierarchical priors (HPs) for the covariance matrix that produce posterior shrinkage toward a specified structure-here we examine shrinkage toward diagonality. We then address the computational difficulties raised by incorporating these priors, and nonconjugate priors in general, into hierarchical models. We apply a combination of approximation, Gibbs sampling (possibly with a Metropolis step), and importance reweighting to fit the models, and compare this hybrid approach to alternative Markov Chain Monte Carlo methods. Our investigation involves three alternative HPs. The first works with the spectral decomposition of the covariance matrix and produces both shrinkage of the eigenvalues toward each other and shrinkage of the rotation matrix toward the identity. The second produces shrinkage of the correlations toward 0, and the third uses a conjugate Wishart distribution to shrink toward diagonality. A simulation study shows that the first two HPs can be very effective in reducing small-sample risk, whereas the conjugate Wishart version sometimes performs very poorly. We evaluate the computational algorithm in the context of a normal nonlinear random-effects model and illustrate the methodology with a logistic random-effects model.

Page Thumbnails

  • Thumbnail: Page 
1254
    1254
  • Thumbnail: Page 
1255
    1255
  • Thumbnail: Page 
1256
    1256
  • Thumbnail: Page 
1257
    1257
  • Thumbnail: Page 
1258
    1258
  • Thumbnail: Page 
1259
    1259
  • Thumbnail: Page 
1260
    1260
  • Thumbnail: Page 
1261
    1261
  • Thumbnail: Page 
1262
    1262
  • Thumbnail: Page 
1263
    1263