Epidemiology Meets Quantum: Statistics, Causality, and Bell's Theorem

• Activity Number: 401
• Type: Invited
• Date/Time: Tuesday, August 11, 2015 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Richard David Gill*
• Companies: Leiden University
• Keywords: Bell's theorem ; quantum nonlocality ; entanglement ; graphical models ; causality ; experimental design

Abstract

I will present a simple proof of Bell's theorem - "quantum mechanics contradicts the classical physical principles of locality and realism". My proof is built around a finite N, probabilistic or statistical version of the Bell inequality; the usual physicist's version is obtained by taking the large N limit. I will explain the many links to thinking about causality in modern statistics (graphical models, counterfactual variables) and other key statistical concepts (randomisation in experimental design, for instance). The new proof also turns out to be extremely useful when engaging with quantum crackpots. Coming up with disproofs of Bell's theorem is the present day version of the perpetuum mobile or squaring the circle. The quantum Randi challenge or QRC (a notion recently invented by Sascha Vongehr) is the challenge of writing an impossible computer program (at least, as long as we do not have quantum computers). 50 years after Bell's theorem, the definitive experiment still has not been done due to statistical design issues! Experimentalists expect to get there, any day now. Reference: http://arxiv.org/abs/1207.5103 (To appear in Statistical Science, special issue on causality).