User Guide

This guide explains the concepts behind geossm’s building blocks and how they fit together. It complements the Quick Start (a single worked example) and the API Reference (full parameter-level reference).

The typical pipeline is:

1. Data preparation – turn a geopandas.GeoDataFrame and a model formula into design matrices organized by space and time.

2. Covariance modeling – build a mesh over the spatial domain and a finite-element (SPDE) approximation of the Matérn covariance on it.

3. State-space modeling – filter, smooth, and estimate a linear-Gaussian state-space model, either directly from system matrices or, for large spatial domains, through the low-rank model that ties the previous two steps together.

• Data preparation
• Covariance models and meshes
• State-space models
• Low-rank state-space models

Development: building this documentation

pip install -e ".[docs]"
sphinx-build -b html docs/source docs/build/html
python -m http.server --directory docs/build/html 8000

See Installation for the full local-build and preview instructions.