Saved in:
Bibliographic Details
Main Author: Osmanagich, Sam
Format: Recurso digital
Language:
Published: Zenodo 2026
Subjects:
Online Access:https://doi.org/10.5281/zenodo.19659242
Tags: Add Tag
No Tags, Be the first to tag this record!
Table of Contents:
  • <p>Geometric patterns identified in spatial data after inspection are difficult to evaluate statistically. <br>When hypotheses are formulated a posteriori, conventional tests can overestimate significance <br>because exploratory choices are not accounted for. This problem is pronounced in small-N spatial <br>point sets, where model flexibility and feature selection strongly influence outcomes.A constrained <br>evaluation framework is applied to assess a posteriori geometric hypotheses in spatial data. The <br>approach limits the geometric degrees of freedom and conditions tests on a fixed set of candidate <br>points. It is intended for situations in which a geometric pattern is first observed and then formally <br>assessed. Point-to-curve deviations are used to compare the observed configuration with alternative <br>spatial and geometric arrangements subject to specified constraints.The framework is demonstrated <br>using a summit landscape in Central Bosnia, where a constrained logarithmic curve pattern has been <br>proposed to link a small set of named summit locations derived from LiDAR data. The observed <br>configuration occupies an extreme position relative to alternative constrained configurations within <br>the defined summit set.The analysis is limited to spatial geometry and does not address origin or <br>interpretation. The contribution is a transparent method for evaluating a posteriori geometric <br>hypotheses in small-N spatial datasets.</p>