• Corpus ID: 247594613

Exact Anytime-valid Confidence Intervals for Contingency Tables and Beyond

@inproceedings{Turner2022ExactAC,
  title={Exact Anytime-valid Confidence Intervals for Contingency Tables and Beyond},
  author={Rosanne Turner and Peter D. Grunwald},
  year={2022}
}
E-variables are tools for retaining type-I error guarantee with optional stopping. We extend E-variables for sequential two-sample tests to general null hypotheses and anytime-valid confidence sequences. We provide implementations for estimating risk difference, relative risk and odds-ratios in contingency tables. 

Figures from this paper

Beyond Neyman-Pearson
TLDR
It is proposed to replace confidence intervals and distributions by the e-posterior , which provides valid post-hoc frequentist uncertainty assessments irrespective of prior correctness: if the prior is chosen badly, e-intervals get wide rather than wrong, suggesting the eThe resulting quasi-conditional paradigm addresses foundational and practical issues in statistical inference.

References

SHOWING 1-10 OF 17 REFERENCES
Safe Tests and Always-Valid Confidence Intervals for contingency tables and beyond
TLDR
Comparison to p-value analysis in simulations and a real-world example show that E-variables, through their flexibility, often allow for early stopping of data collection, thereby retaining similar power as classical methods.
Generic E-Variables for Exact Sequential k-Sample Tests that allow for Optional Stopping
We develop E -variables for testing whether two or more data streams come from the same source or not, and more generally, whether the difference between the sources is larger than some minimal effect
E-values: Calibration, combination, and applications
TLDR
It is demonstrated that e-values are often mathematically more tractable; in particular, in multiple testing of a single hypothesis, e- Values can be merged simply by averaging them, which allows to develop efficient procedures using e- values for testing multiple hypotheses.
Valid sequential inference on probability forecast performance
Probability forecasts for binary events play a central role in many applications. Their quality is commonly assessed with proper scoring rules, which assign forecasts a numerical score such that a
The Language of Betting as a Strategy for Statistical and Scientific Communication
TLDR
A simple betting interpretation of likelihood ratios is called on, which leads to methods that lend themselves to meta-analysis and accounting for multiple testing, and does not encourage the fallacy that probabilistic models imply the existence of unseen alternative worlds.
Uniform, nonparametric, non-asymptotic confidence sequences
TLDR
This paper develops non-asymptotic confidence sequences that achieve arbitrary precision under nonparametric conditions and strengthens and generalizes existing constructions of finite-time iterated logarithm ("finite LIL") bounds.
Confidence sequences for mean, variance, and median.
  • D. Darling, H. Robbins
  • Mathematics
    Proceedings of the National Academy of Sciences of the United States of America
  • 1967
UNIVERSITY OF CALIFORNIA, BERKELEY Communicated by J. Neyman, May 1, 1967 1. Normal Means and Variances.-Let x1,x2,... be independent, identically dis1 n tributed random variables and put ((t) =
Estimation of mixture models
TLDR
A new information projection theory is derived to expand the approximating class to include its information closure, and the existence and uniqueness of a f* in the closure of the convex hull C is identified, such that D ( fpf* ) = infg∈CD fpg , where Dfpg is the Kullback-Leibler divergence.
Safe testing. accepted, pending minor revision, for publication in
  • Journal of the Royal Statistical Society: Series B
  • 2022
Sequential Tests of Statistical Hypotheses.
...
...