Searcharxiv⌕ Search

arXiv subjects

Marian Grendar

Publications and source records attributed to Marian Grendar.

9 recordsLinked to original sources
arXiv2026

Detection coherence of tests

There exist tests calibrated under a null narrower than the one implied by their test statistic -- the detection-null set. The part of the detection-null set not in the null is the test's blind spot. A framework for assessing detection coherence -- whether a test's blind spot is empty -- is introduced, built on new concepts of calibration statistic, detector, detection functional, detection-null set, and blind spot. It is demonstrated that the Wilcoxon--Mann--Whitney, Kruskal--Wallis, Friedman, and logrank tests are detection incoherent as unrestricted tests. Detection incoherent tests may become coherent under restrictions that make the blind spot empty, though domain restriction is not a reliable route to coherence in practice. The Kolmogorov--Smirnov and Zaremba tests, and the Maximum Mean Discrepancy test with a characteristic kernel, are shown to be universally detection coherent. Due to the blind spot, a detection incoherent test misses discoveries when non-rejecting and yields spurious discoveries when rejecting, in both cases regardless of sample size. Detection incoherent tests should be abandoned.

stat.ME↗
arXiv2025

Wilcoxon-Mann-Whitney Test of No Group Discrimination

The traditional WMW null hypothesis $H_0: F = G$ is erroneously too broad. WMW actually tests narrower $H_0: AUC = 0.5$. Asymptotic distribution of the standardized $U$ statistic (i.e., the empirical AUC) under the correct $H_0$ is derived along with finite sample bias corrections. The traditional alternative hypothesis of stochastic dominance is too narrow. WMW is consistent against $H_1: AUC \neq 0.5$, as established by Van Dantzig in 1951.

stat.ME↗
arXiv2008

Generalized Maxwell-Boltzmann, Bose-Einstein, Fermi-Dirac and Acharya-Swamy Statistics and the Polya Urn Model

Generalized probability distributions for Maxwell-Boltzmann, Bose-Einstein and Fermi-Dirac statistics, with unequal source probabilities $q_i$ for each level $i$, are obtained by combinatorial reasoning. For equiprobable degenerate sublevels, these reduce to those given by Brillouin in 1930, more commonly given as a statistical weight for each statistic. These distributions and corresponding cross-entropy (divergence) functions are shown to be special cases of the Pólya urn model, involving neither independent nor identically distributed ("ninid") sampling. The most probable Pólya distribution contains the Acharya-Swamy intermediate statistic.

cond-mat.stat-mech↗
arXiv2007

Conditional Equi-concentration of Types

Conditional Equi-concentration of Types on I-projections is presented. It provides an extension of Conditional Weak Law of Large Numbers to the case of several I-projections. Also a multiple I-projections extension of Gibbs Conditioning Principle is developed. mu-projection variants of the probabilistic laws are stated. Implications of the results for Relative Entropy Maximization, Maximum Probability, Maximum Entropy in the Mean and Maximum Renyi-Tsallis Entropy methods are discussed.

math.PR↗
arXiv2006

The Polya Urn: Limit Theorems, Polya Divergence, Maximum Entropy and Maximum Probability

Sanov's Theorem and the Conditional Limit Theorem (CoLT) are established for a multicolor Polya Eggenberger urn sampling scheme, giving the Polya divergence and the Polya extension to the Maximum Relative Entropy (MaxEnt) method. Polya MaxEnt includes the standard MaxEnt as a special case. The universality of standard MaxEnt - advocated by an axiomatic approach to inference for inverse problems - is challenged, in favor of a probabilistic approach based on CoLT and the Maximum Probability principle.

cond-mat.stat-mech↗
arXiv2004

Gibbs conditioning extended, Boltzmann conditioning introduced

Conditional Equi-concentration of Types on I-projections (ICET) and Extended Gibbs Conditioning Principle (EGCP) provide an extension of Conditioned Weak Law of Large Numbers and of Gibbs Conditioning Principle to the case of non-unique Relative Entropy Maximizing (REM) distribution (aka I-projection). ICET and EGCP give a probabilistic justification to REM under rather general conditions. mu-projection variants of the results are introduced. They provide a probabilistic justification to Maximum Probability (MaxProb) method. 'REM/MaxEnt or MaxProb?' question is discussed, briefly. Jeffreys Conditioning Principle is mentioned.

math-ph↗
arXiv2000

Minimax Entropy and Maximum Likelihood. Complementarity of tasks, identity of solutions

Concept of exponential family is generalized by simple and general exponential form. Simple and general potential are introduced. Maximum Entropy and Maximum Likelihood tasks are defined. ML task on the simple exponential form and ME task on the simple potentials are proved to be complementary in set-up and identical in solutions. ML task on the general exponential form and ME task on the general potentials are weakly complementary, leading to the same necessary conditions. A hypothesis about complementarity of ML and MiniMax Entropy tasks and identity of their solutions, brought up by a special case analytical as well as several numerical investigations, is suggested in this case. MiniMax Ent can be viewed as a generalization of MaxEnt for parametric linear inverse problems, and its complementarity with ML as yet another argument in favor of Shannon's entropy criterion.

math.ST↗
arXiv2000

What is the question that MaxEnt answers? A probabilistic interpretation

The Boltzmann-Wallis-Jaynes' multiplicity argument is taken up and elaborated. MaxEnt is proved and demonstrated to be just an asymptotic case of looking for such a vector of absolute frequencies in a feasible set, which has maximal probability of being generated by a uniform prior generator/pmf.

math-ph↗