What is it about?

Shows that shrinkage of results is a simple consequence of random variation.

Featured Image

Why is it important?

Provides a simple demonstration of the fact that just because forward unbiasedness applies (whereby a statistic is on average equal to the parameter) does not mean that backward unbiasedness applies (whereby a parameter is on average equal to the statistic).

Read the Original

This page is a summary of: Transposed Conditionals, Shrinkage, and Direct and Indirect Unbiasedness, Epidemiology, September 2008, Wolters Kluwer Health,
DOI: 10.1097/ede.0b013e318181b3e3.
You can read the full text:

Read

Resources

Contributors

The following have contributed to this page