Entropy Step Geometry Under Matched Empirical and Synthetic Controls

This repository contains the research article “Entropy Step Geometry Under Matched Empirical and Synthetic Controls”, an empirical study examining whether entropy step geometry reconstructed from multifractal measurements contains structural information capable of distinguishing empirical datasets from synthetic controls. The paper addresses a precise and restricted question: Does entropy step geometry, when resolved in scalar and two-dimensional embeddings under a fixed measurement protocol, encode any structural separation between empirical and synthetic systems? The question is formulated strictly at the measurement level.No dynamics, causality, temporal interpretation, or physical mechanisms are introduced. Contents This archive contains: the full paper in PDF format a Python analysis script reproducing all figures and reported statistics(P21_Entropy_Step_Geometry_EMP_vs_SYN.py) a README file describing scope, methodology, and reproducibility Measurement context All analyses are performed within the Unified Temporal–Measurement Framework (UTMF v5.2), using a fixed, validated, and fully reproducible multifractal measurement protocol. Within this framework: Fractal entropy is reconstructed deterministically from archived multifractal spectra. Entropy updates are represented as geometric step vectors $(\Delta E_1, \Delta E_2)$. All observables are derived deterministically from archived JSON measurement metadata. No assumptions are made about underlying processes, evolution, causality, or temporal ordering. Synthetic datasets are processed under the identical protocol as empirical datasets. The paper operates exclusively on archived measurement outputs and does not recompute or modify raw empirical data. Scope and objective The purpose of this paper is not to explain entropy, but to determine whether entropy step geometry functions as a discriminative observable. Specifically, the paper evaluates whether empirical and synthetic datasets separate under: scalar entropy step magnitude distributions, tail curvature in log–log CCDF space, two-dimensional entropy-step embeddings, directional angle statistics, conditional directional tightening behavior. The central empirical finding is negative but structurally important: No scalar, tail-based, curvature-based, or geometric observable examined here yields a structural separation between empirical and synthetic entropy under a fixed measurement protocol. Entropy step geometry therefore functions as a constrained measurement-level object rather than a diagnostic of dataset origin or generative mechanism. Position within the research programme This article extends the entropy geometry stack established in the Fractal Entropy series and subsequent constraint papers. It builds directly on: Fractal Entropy as a Constrained Geometric Object: Definitions, Bounds, Stability, and Admissible TrajectoriesDOI: https://doi.org/10.5281/zenodo.18489405 Entropy Geometry and Extreme Event Realization in Empirical SystemsDOI: https://doi.org/10.5281/zenodo.18501234 Minimal Admissible Entropy GeometryDOI: https://doi.org/10.5281/zenodo.18529378 Where P20 established the minimal geometric structure sufficient to host bounded entropy behavior, the present paper tests whether that structure differentiates empirical systems from synthetic constructions. It does not introduce new entropy definitions or measurement procedures.It tests the discriminative limits of the existing geometry. What this paper does (and does not do) What it does Compares empirical and synthetic entropy step geometry under identical measurement conditions Identifies structural invariances across heterogeneous dataset classes Tests whether entropy step observables encode dataset origin Establishes entropy step geometry as a measurement-level invariant Defines empirical limits on entropy as a discriminative observable What it does not do Propose a generative or dynamical model Introduce temporal evolution or causal structure Attribute meaning, realism, or authenticity to entropy geometry Treat entropy as a classifier or predictor Tune synthetic data to mimic empirical geometry How to read this paper This article should be read as a constraint test, not as a theory of entropy. It asks: “Does entropy step geometry distinguish empirical from synthetic systems?” and answers: “Under a fixed and validated measurement protocol, it does not.” The result does not imply equivalence of underlying processes.It establishes only that entropy geometry, at the measurement level examined here, does not encode origin. Readers are encouraged to treat the findings as boundary conditions on interpretation rather than explanatory conclusions. Related Works The present study builds directly on a sequence of empirical investigations into multifractal measurement geometry, entropy reconstruction, and asymmetry structure under the Unified Temporal–Measurement Framework (UTMF). Foundational measurement framework The multifractal estimation procedure used throughout the programme is introduced in: A Minimal and Robust Implementation of Multifractal Detrended Fluctuation Analysis (P01) and operationalized at scale in: UTMF-Core (P02) Subset Design and Stability Regimes in MFDFA (P03) These works establish the reproducible measurement protocol underlying all subsequent analyses. Asymmetry geometry and hierarchical structure The relational geometry of multifractal datasets was first formalized in: Fractal Asymmetry Theory (FAT) (P05) and extended through: Emergence Depth (P06) Structural Memory (P07) Emergence Is Confined (P08) Incremental Emergence (P09) These studies characterize hierarchical integration and structural positioning within asymmetry space. Fractal Asymmetry Kernel (FAK) The statistical geometry of admissible asymmetry updates was introduced in: The Fractal Asymmetry Kernel (FAK) (P10) and expanded in: FAK-2: Tail Structure and Energetic Regimes (P11) FAK-3: Contextual Isolation of Extreme Updates (P12) These works establish heavy-tailed but constrained update statistics at the asymmetry level, independent of entropy reconstruction. Entropy geometry Entropy was subsequently reconstructed from multifractal outputs and analyzed in a dedicated series: Empirical Definitions of Fractal Entropy (P14) Bounds and Stability of Fractal Entropy (P15) Fractal Entropy vs Noise (P16) Fractal Entropy as a Constrained Trajectory (P17) Fractal Entropy as a Constrained Geometric Object (P18) These works establish bounded entropy support, admissible trajectory structure, and dataset-specific variability within a shared geometric domain. The role of extreme asymmetry realization in entropy space was addressed in: Entropy Geometry and Extreme Event Realization (P19) and the minimal geometric structure sufficient to host all observed entropy behavior was identified in: Minimal Admissible Entropy Geometry (P20) Position of the present paper The present study (P21) extends this line by shifting focus from entropy support and trajectory admissibility to the geometry of entropy actualization steps. Unlike P16, which compares entropy magnitudes under synthetic controls, this paper resolves entropy increments in two-dimensional embedding and examines: step magnitude distributions, tail curvature structure, directional geometry, and conditional magnitude–direction coupling. It therefore addresses a narrower but more structural question: Does entropy step geometry encode a discriminative signature separating empirical and synthetic datasets under a fixed measurement protocol? The empirical answer provided here is negative, thereby completing the measurement-level characterization of entropy geometry under UTMF. Citation If you use this archive, please cite both: the Zenodo record itself, and the accompanying paper included herein.