We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Statistics > Methodology

Title: High Dimensional Linear Regression via the R2-D2 Shrinkage Prior

Abstract: We propose a new class of priors for linear regression, the R-square induced Dirichlet Decomposition (R2-D2) prior. The prior is induced by a Beta prior on the coefficient of determination, and then the total prior variance of the regression coefficients is decomposed through a Dirichlet prior. We demonstrate both theoretically and empirically the advantages of the R2-D2 prior over a number of common shrink- age priors, including the Horseshoe, Horseshoe+, and Dirichlet-Laplace priors. The R2-D2 prior possesses the fastest concentration rate around zero and heaviest tails among these common shrinkage priors, which is established based on its marginal density, a Meijer G-function. We show that its Bayes estimator converges to the truth at a Kullback-Leibler super-efficient rate, attaining a sharper information theoretic bound than existing common shrinkage priors. We also demonstrate that the R2-D2 prior yields a consistent posterior. The R2-D2 prior permits straightforward Gibbs sampling and thus enjoys computational tractability. The proposed prior is further investigated in a mouse gene expression application.
Subjects: Methodology (stat.ME)
Cite as: arXiv:1609.00046 [stat.ME]
  (or arXiv:1609.00046v2 [stat.ME] for this version)

Submission history

From: Yan (Dora) Zhang [view email]
[v1] Wed, 31 Aug 2016 21:16:51 GMT (456kb,D)
[v2] Sun, 10 Sep 2017 15:13:29 GMT (1198kb,D)

Link back to: arXiv, form interface, contact.