Family-wise control implies false-discovery control
Theorem taught in High-dimensional analysis, part of Statistical learning
Family-wise control implies false-discovery control
| Type | Theorem |
| Taught in | High-dimensional analysis |
| Part | Statistical learning |
When all null hypotheses are true (m_0=m), \mathrm{FDR}=\mathrm{FWER}. In general \mathrm{FDR}\le\mathrm{FWER}.
Used in
- Taught in High-dimensional analysis, where the derivation and examples are.
- All 9 questions on High-dimensional analysis
See also
Null and alternative hypotheses · Rank bound · Benjamini–Hochberg · Benjamini–Hochberg controls the false discovery rate · Why classical methods fail when p > n · Rank of the sample covariance · Regression when p > n · Where high-dimensional columns come from