The empirical Bayes normal means (EBNM) model is important to many areas of statistics, including (but not limited to) multiple testing, wavelet denoising, multiple linear regression, and matrix factorization. There are several existing software packages that can fit EBNM models under different prior assumptions and using different algorithms; however, the differences across interfaces complicate direct comparisons. Further, a number of important prior assumptions do not yet have implementations. Motivated by these issues, we developed the R package ebnm, which provides a unified interface for efficiently fitting EBNM models using a variety of prior assumptions, including nonparametric approaches. In some cases, we incorporated existing implementations into ebnm; in others, we implemented new fitting procedures with a focus on speed and numerical stability. To demonstrate the capabilities of the unified interface, we compare results using different prior assumptions in two extended examples: the shrinkage estimation of baseball statistics; and the matrix factorization of genetics data (via the new R package flashier). In summary, ebnm is a convenient and comprehensive package for performing EBNM analyses under a wide range of prior assumptions.
View on arXiv