Spatstat: Spatial Point Pattern Analysis and Modeling Toolbox

Spatstat is an open-source R package providing a toolbox for analyzing spatial point patterns. It contains over 3000 functions for exploratory data analysis, model-fitting, simulation, and formal inference for point patterns in two or three dimensions, on linear networks, and in space-time. The package was authored by Adrian Baddeley.

Use Cases

• Fitting parametric point process models like Poisson or Gibbs processes using functions such as ppm() and kppm()
• Performing formal hypothesis tests for randomness or covariate effects using methods like Monte Carlo or Kolmogorov-Smirnov tests
• Generating simulation envelopes and diagnostic plots like Q-Q plots for model validation
• Analyzing marked point patterns using mark correlation functions and segregation indices

Strengths

• Contains over 3000 functions for a wide range of spatial analyses
• Supports multiple complex data types including point patterns on linear networks and space-time patterns
• Implements a variety of formal hypothesis tests and model-fitting methods similar to generalized linear models

Limitations

• Column-level documentation is absent; field semantics must be inferred after download
• Row count is unknown, which may limit suitability assessment
• Last update date is unknown; freshness unverified

Provenance

Source

Adrian Baddeley