AI & ML
A linear programming approach to finite copula interpolation
arXiv stat.TH0 views
arXiv:2611.14008v1 Announce Type: new Abstract: We consider exact and interval-valued interpolation of multivariate copulas at arbitrary finite configurations in the unit cube. We first characterize when the prescribed data are compatible with a quasi-copula and derive explicit extremal quasi-copula envelopes. The copula interpolation problem is thereby reduced to a copula sandwich problem between these envelopes. Restricting them to the canonical rectangular grid generated by the data coordinates yields a finite linear feasibility system whose elementary-volume constraints are encoded by a cell--vertex incidence matrix. Farkas' lemma then identifies the criterion of Omladi\v{c} and Stopar with an explicit certificate of infeasibility. Although this criterion is indexed by infinitely many formal cell combinations, we show that it suffices to test finitely many vertices of a sign-polytopal subdivision. We also extend the construction to $k$-increasing discrete quasi-copulas for $2\leq k\leq d$.
Source:https://arxiv.org/abs/2611.14008PreviousAsymptotic Bayes Optimality of Sparse Multiple Testing Procedures under One-Factor DependenceNextOn the geometry of Gaussian Mean Shift Detection
Related posts
Hitting 200 Strokes: What the Signature Maker Does at Its Stroke Limit
0dev.to
A Lightweight, Open-Source Network Digital Twin Blueprint for IoT-Edge Environments
0arXiv cs.NI
Bridge: Optimizing Collective Communication Schedules in Reconfigurable Networks with Reusable Subrings
0arXiv cs.NI
SplitCom: Communication-efficient Split Federated Fine-tuning of LLMs via Temporal Compression
0arXiv cs.NI
Programmable GPU-Accelerated dApps for Real-Time Edge Inference and ISAC on NVIDIA Aerial Testbed
0arXiv cs.NI
PSL-RGHA: Parallel Split Learning with Reference Gradient Half-Space Alignment for Mitigating Gradient Divergence
0arXiv cs.NI