Searcharxiv⌕ Search
Home/Archive/arXiv

arXiv · 2610.09951

Overlapped We Stand, Separated We Fall: Software Approaches to Diagnosing Separation Phenomena in Categorical Response Models

Thomas Rusch·Lukas Sablica·Kurt Hornik
View original source ↗Read source document ↗Export metadata ↓

Abstract

In statistical models for categorical responses, including binary, nominal and ordinal responses, a phenomenon called ``separation'' may occur which can render maximum likelihood estimate nonexistent and/or nonunique. Based on Sablica et al. (2026)'s general characterization of separation phenomena via structure vectors, in this article we discuss computational, numerical and practical aspects of separation phenomena in different software packages, including consequences, remedies and ways of diagnosing them. We further introduce the R package divoRce which provides the most comprehensive software suite to diagnose separation in existence. It allows to check for separation, to distinguish between types of separation, characterize the recession cone and identify variables and observations contributing to separation for almost any categorical response model in the literature, including models with baseline-category link, cumulative link, adjacent-category link and sequential link specifications. Diagnostics are based on methods for computational geometry and linear programming and can be done in floating-point or rational arithmetic with different solvers. The functionality is easily extendible by wrappers and S3 methods to accommodate implementations in third-party packages or of new categorical response models. We discuss our implementation in detail and show our software in action in a wide range of applications. We recommend to incorporate our checks for separation and existence of the MLE as a diagnostic routine in every categorical data analysis both in third-party software implementations and as part of a standard model diagnostic workflow.

Explore related subjects

stat.COstat.AP

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Thomas Rusch, Lukas Sablica, Kurt Hornik. 2026-10-07. Overlapped We Stand, Separated We Fall: Software Approaches to Diagnosing Separation Phenomena in Categorical Response Models. https://arxiv.org/abs/2610.09951

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

NExt-LMM: a two-stage LMM framework with near-exact variance components for genome-wide association studies

Linear mixed models (LMMs) are widely used in genome-wide association studies (GWAS) to account for population stratification and cryptic relatedness. However, their practical application remains computationally challenging because parameter estimation and association testing require large-scale operations on genetic similarity matrices (GSMs). Here, we present NExt-LMM, a two-stage framework for scalable LMM-based GWAS. In the first stage, NExt-LMM obtains a preliminary estimate of a shared variance ratio under the null model. In the second stage, it constructs the corresponding covariance matrix and accelerates downstream inference through a Hierarchical Off-Diagonal Low-Rank (HODLR) representation. Specifically, NExt-LMM exploits the low-rank structure of off-diagonal covariance blocks to enable substantially faster matrix factorization and inverse-related computations than conventional dense decompositions. It then reuses the resulting HODLR-based covariance representation to accelerate maximum-likelihood-based inference under the shared variance-ratio setting. We further establish error bounds that account for both sources of approximation in the framework: the preliminary variance-ratio estimate and the HODLR approximation. In particular, we show that, conditional on a fixed variance ratio, the NExt-LMM estimator can be made arbitrarily close to the exact shared-variance estimator as the matrix approximation tolerance approaches zero. \AAA{Numerical experiments show that NExt-LMM improves computational efficiency while maintaining high accuracy relative to existing methods, and that the size of this improvement is governed by the numerical rank retained at the tolerance required for calibrated association inference.} We provide an open-source Python implementation at https://github.com/ZhibinPU/NExt-LMM.

stat.CO↗
arXiv2026

On the Convergence of Wasserstein Gradient Descent for Sampling

This paper studies the optimization of the KL functional on the Wasserstein space of probability measures, and develops a sampling framework based on Wasserstein gradient descent (WGD). We identify two important subclasses of the Wasserstein space for which the WGD scheme is guaranteed to converge, thereby providing new theoretical foundations for optimization-based sampling methods on measure spaces. For practical implementation, we construct a particle-based WGD algorithm in which the score function is estimated via score matching. Through a series of numerical experiments, we demonstrate that WGD can provide good approximation to a variety of complex target distributions, including those that pose substantial challenges for standard MCMC and parametric variational Bayes methods. These results suggest that WGD offers a promising and flexible alternative for scalable Bayesian inference in high-dimensional or multimodal settings.

stat.CO↗
arXiv2026

postshock: An R Package for Donor-Adjusted Forecasting After Structural Shocks

We present the postshock R package, which implements and extends a donor-based framework for forecasting when a structural shock is known and the target response of interest is not yet observed. The package estimates shock effects from historical donor episodes, balances donors using specified matching features, and transfers the resulting adjustment to a target-series forecast. It provides integrated workflows for conditional mean forecasting through ARIMA and ARIMAX models and conditional variance forecasting through GARCH-X models. Additional functionality includes structured donor pools, control-shock regressors, automated GARCH-X order selection, processed data objects, and reproducible empirical workflows.

stat.CO↗