JointShrink

Multivariate adaptive shrinkage uses shared patterns across conditions to estimate effects and their uncertainty. JointShrink provides the Gaussian summary-statistic method of Urbut et al. (2019) in Python, together with a native multivariate Student-t likelihood and direct outcome-regression extensions.

pip install jointshrink

Analysis	Start with
Supplied effects and standard errors	Gaussian summaries
Heavy-tailed summary errors	Student-t summaries
Learn shared patterns	Covariance learning
Outcomes and predictors	Direct regression
Event times, categories or excess zeros	Regression families
Missing summary conditions or larger inputs	Prior reuse and batches
Function arguments and result methods	Grouped API

The tutorials use constructed inputs. Gaussian summary inference conditions on fitted prior parameters; direct regression uses posterior sampling. Numerical agreement checks do not establish interval calibration or convergence in every analysis. See methods and limits.

Installation · Reference manual · Numerical comparisons

Citation

See the package paper and BibTeX citation.

Acknowledgements

Computations were performed on the Arseven Computing Cluster, Department of Statistics, Texas A&M University.

This work received no grant funding.