遇见数据集

Precision Cascade Reproducibility Archive for "Predictive Synthesis of Earth-Analog Occurrence Rates Using Structural Covariance" (v3.0.0, Invariant-Audited)

收藏
Zenodo2026-04-16 更新2026-05-26 收录
官方服务:

资源简介:

This archive implements a structural-covariance synthesis framework (Precision Cascade Engine), providing a complete, deterministic, and self-auditing reproducibility system for the manuscript: “Predictive Synthesis of Earth-Analog Occurrence Rates Using Structural Covariance Across Transit, Radial-Velocity, Microlensing, and Direct-Imaging Surveys”. The system is designed to deterministically reproduce and audit the full statistical pathway from heterogeneous exoplanet occurrence-rate measurements to predictive synthesis. All numerical results are generated internally from: declared occurrence-rate measurements provenance metadata across stellar models, atmospheric physics, detection pipelines, and calibration frameworks structural-layer encodings correlation-matrix construction rules covariance synthesis systematic correction procedures The archive reproduces the primary results: Earth-analog occurrence rate: 2.2% ± 0.1% Effective degrees of freedom: 20.1 (from 36 measurements) Roman Space Telescope prediction: 11.2 ± 0.2 planets per 1000 stars Key features: Fully deterministic execution (numerically consistent within machine precision) Single self-contained archive with no hidden dependencies Explicit provenance → correlation → covariance construction Inverse-covariance synthesis (Precision Cascade estimator) Effective degrees-of-freedom computation Executable regeneration pipeline from a single scripted command Machine-readable outputs for all results and diagnostics Structural sensitivity tests (method removal, correlation perturbation, systematics scaling) Internal validation against golden outputs Checksum closure with ROOT_HASH for integrity verification Self-auditing framework with PASS/FAIL governance Explicit invariant diagnostics demonstrating irreducible relationships within the model Transparent disclosure of all residuals between literal computation and reported benchmarks The archive is designed to: enable full independent replication support adversarial testing of structural assumptions provide a complete audit trail from raw inputs to final predictions demonstrate reproducibility-first scientific inference across heterogeneous datasets The archived version corresponds to v3.0.0 (“invariant-audited”), identical to the version used to generate all results reported in the manuscript. This work establishes a structurally transparent, predictive framework for cross-method synthesis in exoplanet demographics. Keywords exoplanets, occurrence rate, Earth analogs, Roman Space Telescope, covariance, statistical inference, reproducibility, astrophysics, scientific computing, precision cascade, auditability

提供机构:
Zenodo
创建时间:
2026-04-16
二维码
社区交流群
二维码
科研交流群
商业服务