Contributor
Jeffery Sauer

PySAL ESDA Enhancements: Local join count and LOSH statistics


Mentors
Stefanie Lumnitz, Dani Arribas-Bel, Levi John Wolf
Organization
NumFOCUS

The goal of this project is to add several recently developed spatial estimators to the exploratory spatial data analysis (esda) submodule of PySAL, the Python Spatial Analysis Library. This project will allow researchers to easily deploy these estimators in existing spatial workflows. Specifically, this project will contribute implementation, docstrings, tests, and example notebooks for (1) bivariate local join count statistics, (2) multivariate local join count statistics, and (3) local spatial heteroskedasticity (LOSH) statistics. In the first phase of this project, each estimator will be reviewed and pseudo-coded to identify areas of optimization. In the second phase, the estimators will be implemented with tests, with performance assessed against spatial datasets of different sizes. In the final phase, example notebooks will be drafted and polished to a quality ready for external (e.g. workshop) use.