This will take the output of either mldest() or sldest(), shrink the Fisher-z transformed correlation estimates using ash() (Stephens, 2017; Dey and Stephens, 2018), then return the corresponding correlation estimates. You can obtain estimates of r^2 by just squaring these estimates.

ldshrink(obj, ...)

Arguments

obj

An object of class lddf, usually created using either mldest() or sldest().

...

Additional arguments to pass to ash().

Value

A correlation matrix.

References

  • Stephens, Matthew. "False discovery rates: a new deal." Biostatistics 18, no. 2 (2017): 275-294.

  • Dey, Kushal K., and Matthew Stephens. "CorShrink: Empirical Bayes shrinkage estimation of correlations, with applications." bioRxiv (2018): 368316.

Author

David Gerard