Computing the Statistics#
The acm package provides several clustering statistics estimators that can be used to analyze galaxy clustering beyond the traditional 2-point correlation function. These statistics are designed to capture non-Gaussian information in the density field at small scales.
Available Statistics#
The acm.estimators module contains implementations of various clustering statistics:
Galaxy Clustering Statistics#
The main clustering statistics available in acm.estimators.galaxy_clustering include:
Density-Split Statistics: Computes auto- and cross-correlations in different density quantiles
See the Density-Split documentation for detailed usage
Two-Point Correlation Function (TPCF): Standard clustering measurement
Power Spectrum: Fourier-space clustering measurement
Bispectrum: Three-point statistics in Fourier space
Minkowski Functionals: Geometric and topological measures of the density field
Minimum Spanning Tree (MST): Graph-based clustering statistics
See the MST documentation for details
k-Nearest Neighbors (kNN): Distance-based clustering measurements
Computing Statistics#
Each estimator has its own class and interface. The general workflow is:
Initialize the estimator with appropriate parameters
Assign data positions to the estimator
Compute the desired statistic
Save or analyze the results
Tip
Example notebooks showing how to compute statistics like density-split are available in the Tutorials section.
Storage Conventions#
After computing statistics, they should be stored following the conventions described in the Data Storage section. This ensures compatibility with the emulator training pipeline.
See also
For details on specific statistics, see the Statistics section
For information on how to integrate new statistics, see the Projects section
API Reference#
For detailed API documentation of all available estimators, see the API Reference.