zwulab.r-universe.dev

[zwulab] GFisher 0.3.1

zheyangwu@wpi.edu (Zheyang Wu) — Wed, 02 Sep 2026 21:30:17 GMT

This R package provides accurate and computationally efficient methods for computing p-values for a general family of Fisher-type statistics (GFisher), including Fisher's combination, Good's statistic, and Lancaster's statistic. It supports flexible weighting schemes and an omnibus procedure that adapts the weights and degrees of freedom to the data. The package includes p-value calculation methods based on moment-ratio matching and joint distribution approximations. It offers both pure R implementations and optimized C++ code via RcppArmadillo for improved performance on large-scale data. For fast calculations under independence, the optional 'coga' package is required for functions with the '_ind' suffix.

[zwulab] Qapprox 0.2.0

hzhang@wpi.edu (Hong Zhang) — Sun, 09 Aug 2026 17:51:58 GMT

Calculates the right-tail probability of quadratic forms of Gaussian variables using the skewness-kurtosis ratio matching method, modified Liu-Tang-Zhang method and Satterthwaite-Welch method. The technical details can be found in Hong Zhang, Judong Shen and Zheyang Wu (2020) .

[zwulab] DPComb 1.0

gonzalo.contador@usm.cl (Gonzalo Contador) — Sun, 09 Aug 2026 17:51:58 GMT

Provides tools for performing p-value combination tests with discrete input p-values. These tests combine significance evidence derived from independent discrete statistics to test a global null hypothesis, which is defined by the specified null distribution(s) of these discrete statistics. The testing procedure involves two main steps: (1) Wasserstein Adjustment: Each component of the combination statistic is replaced by an adjusted Z statistic. This adjustment, based on the minimum Wasserstein distance, preserves the discrete nature of the original statistics while better aligning them with their counterparts under continuity. (2) Calculation of the Significance of the Combination Statistic: A continuous distribution that optimally matches the discrete distribution of the combination statistic is obtained, and the testing p-value for the global null hypothesis is computed. The first step is analogous to Lancaster's approach but is generalized based on Wasserstein optimization. The second step allows for asymptotic control of Type I error with higher statistical power. The package implements several p-value combination methods, including Fisher’s, Pearson’s, George’s, Stouffer’s, and Edgington’s methods. The individual tests to be combined can be right-sided, left-sided, or two-sided, and can be based on binomial, Poisson, hypergeometric, noncentral hypergeometric, negative binomial, or geometric distributions, or a mixture of them. The underlying methodology and its foundations are described in the following references: Contador, Gonzalo and Wu, Zheyang (2025). A minimum Wasserstein distance approach to Fisher's combination of independent, discrete p-values. Scandinavian Journal of Statistics, 52(3), 1281-1300. Contador, Gonzalo and Wu, Zheyang (2026). Optimal Adjustment and Combination of Independent Discrete p-Values. Under revision at the Journal of Computational and Graphical Statistics. Lancaster, HO (1949). The combination of probabilities arising from data in discrete distributions. Biometrika, 36(3/4), 370-382. .

[zwulab] SetTest 0.3.0

hzhang@wpi.edu (Hong Zhang) — Sun, 09 Aug 2026 17:51:57 GMT

It provides cumulative distribution function (CDF), quantile, p-value, statistical power calculator and random number generator for a collection of group-testing procedures, including the Higher Criticism tests, the one-sided Kolmogorov-Smirnov tests, the one-sided Berk-Jones tests, the one-sided phi-divergence tests, etc. The input are a group of p-values. The null hypothesis is that they are i.i.d. Uniform(0,1). In the context of signal detection, the null hypothesis means no signals. In the context of the goodness-of-fit testing, which contrasts a group of i.i.d. random variables to a given continuous distribution, the input p-values can be obtained by the CDF transformation. The null hypothesis means that these random variables follow the given distribution. For reference, see [1]Hong Zhang, Jiashun Jin and Zheyang Wu. "Distributions and power of optimal signal-detection statistics in finite case", IEEE Transactions on Signal Processing (2020) 68, 1021-1033; [2] Hong Zhang and Zheyang Wu. "The general goodness-of-fit tests for correlated data", Computational Statistics & Data Analysis (2022) 167, 107379.

[zwulab] TFisher 0.2.0

hzhang@wpi.edu (Hong Zhang) — Sun, 09 Aug 2026 17:51:56 GMT

We provide the cumulative distribution function (CDF), quantile, and statistical power calculator for a collection of thresholding Fisher's p-value combination methods, including Fisher's p-value combination method, truncated product method and, in particular, soft-thresholding Fisher's p-value combination method which is proven to be optimal in some context of signal detection. The p-value calculator for the omnibus version of these tests are also included. For reference, please see Hong Zhang and Zheyang Wu. "TFisher Tests: Optimal and Adaptive Thresholding for Combining p-Values", submitted.