Searcharxiv⌕ Search

arXiv subjects

Sanat Sarkar

Publications and source records attributed to Sanat Sarkar.

4 recordsLinked to original sources

False Discovery Rate Control in Ordered Sequential Testing: Rejection-Count and Fallback Procedures

We study false discovery rate (FDR) control for a fixed family of hypotheses tested sequentially in a prespecified order. Each decision uses only the current and preceding $p$-values, is not subsequently revised, and permits continued testing after non-rejections. We distinguish generalized fixed-sequence procedures based on preceding rejection counts from fallback-type procedures that additionally use recursively transferred testing levels. For rejection-count procedures, we study a position-dependent calibration under independence and a linear rule calibrated through null marginal distribution functions under positive left-tail dependence under the null. We clarify their connections to existing fixed-sequence and online FDR methods. The fallback construction combines the total number of preceding discoveries with level transfer along consecutive rejections, allowing thresholds to reflect the arrangement of earlier outcomes. We establish its FDR control under history-based positive regression dependence and, with a position-specific correction, under arbitrary dependence. For all four constructions, the guarantees extend to every deterministic testing time within the planned horizon. Gaussian simulations assess the effects of signal placement, dependence, and allocation weights. They demonstrate power gains over selected benchmarks while identifying the conservativeness of the arbitrary-dependence correction.

stat.ME↗

A Family-Level Graphical Gatekeeping Procedure with Iterative Retesting

Confirmatory clinical trials often involve multiple objectives grouped into families according to scientific priorities. At the intersection of graphical multiple testing and gatekeeping methods, we develop a Bonferroni-based family-level graphical procedure with partial-rejection-based transfers and iterative re-testing. Families are tested sequentially within each stage, using fixed component tests and a prespecified transition matrix. The updating rule combines current-stage forward contributions based on current family levels with previous-stage backward contributions based on initial allocations. We establish strong global familywise error rate control at a prespecified level under arbitrary dependence. The construction permits backward feedback while retaining fixed component tests and transition coefficients, distinguishing its implementation from forward-only family-level procedures and the specified superchain algorithm. Simulations demonstrate gains over the corresponding No-retest procedure, while identifying an efficiency cost relative to a hypothesis-level graphical comparator with matched initial allocations and edge weights. Clinical illustrations show how the procedure implements family-level priorities and feedback, and an extension accommodates a two-layer structure with multiple families per layer.

stat.ME↗

Large-Scale Multiple Testing for Matrix-Valued Data under Double Dependency

High-dimensional inference based on matrix-valued data has drawn increasing attention in modern statistical research, yet not much progress has been made in large-scale multiple testing specifically designed for analysing such data sets. Motivated by this, we consider in this article an electroencephalography (EEG) experiment that produces matrix-valued data and presents a scope of developing novel matrix-valued data based multiple testing methods controlling false discoveries for hypotheses that are of importance in such an experiment. The row-column cross-dependency of observations appearing in a matrix form, referred to as double-dependency, is one of the main challenges in the development of such methods. We address it by assuming matrix normal distribution for the observations at each of the independent matrix data-points. This allows us to fully capture the underlying double-dependency informed through the row- and column-covariance matrices and develop methods that are potentially more powerful than the corresponding one (e.g., Fan and Han (2017)) obtained by vectorizing each data point and thus ignoring the double-dependency. We propose two methods to approximate the false discovery proportion with statistical accuracy. While one of these methods is a general approach under double-dependency, the other one provides more computational efficiency for higher dimensionality. Extensive numerical studies illustrate the superior performance of the proposed methods over the principal factor approximation method of Fan and Han (2017). The proposed methods have been further applied to the aforementioned EEG data.

stat.ME↗

Adaptive Controls of FWER and FDR Under Block Dependence

Often in multiple testing, the hypotheses appear in non-overlapping blocks with the associated $p$-values exhibiting dependence within but not between blocks. We consider adapting the Benjamini-Hochberg method for controlling the false discovery rate (FDR) and the Bonferroni method for controlling the familywise error rate (FWER) to such dependence structure without losing their ultimate controls over the FDR and FWER, respectively, in a non-asymptotic setting. We present variants of conventional adaptive Benjamini-Hochberg and Bonferroni methods with proofs of their respective controls over the FDR and FWER. Numerical evidence is presented to show that these new adaptive methods can capture the present dependence structure more effectively than the corresponding conventional adaptive methods. This paper offers a solution to the open problem of constructing adaptive FDR and FWER controlling methods under dependence in a non-asymptotic setting and providing real improvements over the corresponding non-adaptive ones.

stat.ME↗