Eikosstatistics & data science

Advanced & Applied Topics · High-Dimensional Statistics & Sparsity

FDR, multiple testing, and error control in high dimensions

Public introduction · full guide in Eikos

Introduction

What this guide covers

Introduction Statistical Decision Theory

Overview

Topics covered

  • Introduction
  • Multiple Inference and Error Control in High Dimensions
  • The Problem of High Dimensional Multiplicity
  • Classification and Taxonomy of Error Rates
  • 2.1 Theoretical Error Rates
  • One-Step Methods: Bonferroni and Šidák Correction

Related concepts

Main connections

  • Multiple testing & FWER/FDR
  • p-value & significance
  • Hypothesis testing
  • Linear algebra & matrices
  • Bayes' theorem & priors
  • Calculus & differentiation

Compatible notation

Mathematical sample in LaTeX

The expression is rendered accessibly, with its LaTeX source preserved below.

P(To commit at least a false positive)=1−P(Do not make any false positives)=1−(1−α)m
View LaTeX sourceP(\text{To commit at least a false positive}) = 1 - P(\text{Do not make any false positives}) = 1 - (1 - \alpha)^m