Performance of Test Supermartingale Confidence Intervals for the Success Probability of Bernoulli Trials

  title={Performance of Test Supermartingale Confidence Intervals for the Success
 Probability of Bernoulli Trials},
  author={Peter Wills and Emanuel Knill and Kevin J. Coakley and Yanbao Zhang},
  journal={arXiv: Statistics Theory},
Given a composite null hypothesis H, test supermartingales are non-negative supermartingales with respect to H with initial value 1. Large values of test supermartingales provide evidence against H. As a result, test supermartingales are an effective tool for rejecting H, particularly when the p-values obtained are very small and serve as certificates against the null hypothesis. Examples include the rejection of local realism as an explanation of Bell test experiments in the foundations of… 

Figures from this paper

Sample-Efficient Device-Independent Quantum State Verification and Certification

Aleksandra Gočanin,1 Ivan Šupić,2, 3 and Borivoje Dakić4, 5 1Faculty of Physics, University of Belgrade, Studentski Trg 12-16, 11000 Belgrade, Serbia 2CNRS, LIP6, Sorbonne Université, 4 place



Appendix to "Accardi contra Bell (cum mundi) : The Impossible Coupling"

It is shown that the randomised design of the Aspect experiment closes a loophole for local realism, and a supermartingale version of the classical Bernstein (1924) inequality guaranteeing a not-heavier-than-Gaussian tail of the distribution of a sum of bounded superMartingale differences.

Asymptotically optimal data analysis for rejecting local realism

Reliable experimental demonstrations of violations of local realism are highly desirable for fundamental tests of quantum mechanics. One can quantify the violation witnessed by an experiment in terms

Interval Estimation for a Binomial Proportion

We revisit the problem of interval estimation of a binomial proportion. The erratic behavior of the coverage probability of the stan- d ardWaldconfid ence interval has previously been remarkedon in

Test Martingales, Bayes Factors and p-Values

A nonnegative martingale with initial value equal to one measures evidence against a probabilistic hypothesis. The inverse of its value at some stopping time can be interpreted as a Bayes factor. If

Loopholes in Bell inequality tests of local realism

Bell inequalities are intended to show that local realist theories cannot describe the world. A local realist theory is one where physical properties are defined prior to and independent of

Efficient quantification of experimental evidence against local realism

An efficient protocol is described for computing bounds from any set of Bell inequalities for any number of parties, measurement settings, or outcomes and can be applied to tests of other properties that are witnessed by linear inequalities.

Time, Finite Statistics, and Bell's Fifth Position

In this contribution to the 2002 Vaxjo conference on the foundations of quantum mechanics and probability, I discuss three issues connected to Bell’s theorem and Bell-CHSH-type experiments: time and

Significant-Loophole-Free Test of Bell's Theorem with Entangled Photons.

A Bell test is reported that closes the most significant of loopholes that provide loopholes for a local realist explanation of quantum mechanics, using a well-optimized source of entangled photons, rapid setting generation, and highly efficient superconducting detectors.

Probability inequalities for sum of bounded random variables

Abstract Upper bounds are derived for the probability that the sum S of n independent random variables exceeds its mean ES by a positive number nt. It is assumed that the range of each summand of S