Hyperbolic statistical inference for Treatment Effects with Circular biomarker of astigmatism

TL;DR

A novel two-sample testing framework for circular data grounded in hyperbolic geometry is proposed, and permutation-based tests for the common concentration case and bootstrap-based procedures for unequal concentrations are developed.

Abstract

Circular biomarkers arise naturally in many biomedical applications, particularly in ophthalmology, where angular measurements such as astigmatism are routinely recorded. Similar directional variables also occur in the study of human body rotations, including movements of the hand, waist, neck, and lower limbs. Motivated by a clinical dataset comprising angular measurements of astigmatism induced by two cataract surgery procedures, we propose a novel two-sample testing framework for circular data grounded in hyperbolic geometry. Assuming von Mises distributions with either common or group-specific concentration parameters, we embed the corresponding parameter spaces into the Poincaré disk, an open unit disk endowed with the Poincaré metric.Under this construction, each von Mises distribution is mapped uniquely to a point in the Poincaré disk, yielding a continuous geometric representation that preserves the intrinsic structure of the parameter space. This embedding enables direct comparison of group distributions via hyperbolic distances, leading to natural and interpretable test statistics. We develop permutation-based tests for the common concentration case and bootstrap-based procedures for unequal concentrations. Extensive simulation studies demonstrate stable empirical size, strong consistency, and superior asymptotic power compared with existing competing methods. The proposed methodology is illustrated through a detailed analysis of the cataract surgery dataset, including a clinically informed restructuring of the original observations. The results highlight the practical advantages of incorporating hyperbolic geometry into the analysis of circular biomedical data and underscore the potential of geometry-aware inference for directional biomarkers.

Figures (6)

• Figure 1: Mapping of the von Mises parameters to Poincaré disk
• Figure 2: Empirical size at $5\%$ level of the proposed permutation-based test under the null hypothesis for sample sizes n = 20, 50, 100, and 200 and equal concentration parameters $\kappa_1 = \kappa_2 \in \{1, 1.5, 3.0\}$, demonstrating accurate size control across settings.
• Figure 3: Empirical power curves under the alternative $\mu_2 \in [0, 2\pi)$ with $\mu_1 = 0$ for sample sizes $n = 20, 50, 100,$ and 200 and equal concentration parameters $\kappa_1 = \kappa_2 \in\{1, 1.5, 3.0\}$. Panels (a)–(c) show the power curves, while panels (d)–(f) display the corresponding empirical power differences with the Z-test of Biswas_2016.
• Figure 4: Empirical size under the null hypothesis $\mu_1 = \mu_2 = \mu_2 \in [0,\pi)$ for sample sizes n = 20, 50, 100, and 200, assuming unequal concentration parameters. Panels (a) and (b) correspond to $(\kappa_1,\kappa_2) =$ (1.5, 3.0) and (1.48, 1.50), respectively.
• Figure 5: Empirical power curves under the alternative $\mu_2 \in [0,\pi)$ with $\mu_1 = 0$ for sample sizes n = 20, 50, 100, 200, assuming unequal concentration parameters. Panels (a) and (b) correspond to ($(\kappa_1,\kappa_2) = (1.5, 3.0)$) and (1.48, 1.50), respectively, while panels (c) and (d) show the corresponding empirical power differences with the existing W-test by biswas2016comparison, respectively.

Theorems & Definitions (9)

• lemma 1: Uniqueness of Minimizer
• proof
• lemma 2: Projection along zero direction
• proof
• Theorem 1: Consistency of the Poincaré Distance-Based Permutation Test
• proof
• proof
• proof
• proof