SpaceTimeBSS-package: Blind Source Separation for Multivariate Spatio-Temporal Data

In SpaceTimeBSS: Blind Source Separation for Multivariate Spatio-Temporal Data

Blind Source Separation for Multivariate Spatio-Temporal Data

Description

Simultaneous/joint diagonalization of local autocovariance matrices to estimate spatio-temporally uncorrelated random fields.

Details

Package:	SpaceTimeBSS
Type:	Package
Version:	0.4-0
Date:	2024-05-29
License:	GPL (>= 2)

Solving the second order blind source separation problem for multivariate space-time random fields. The random fields can be irregular in space but must be regular in time.

The main function of this package is:

stbss

This function computes local autocovariance matrices. The considered temporal lags are integer numbers and the spatial lags are defined by spatial kernel functions. Then, these local autocovariance matrices and the sample covariance are simultaneously/jointly diagonalized.

The package also contains a 9-variate dataset of deseasonalized weekly climate and meteorological measurements from the Italian Veneto region between 2000 and 2022 meteo_veneto.

Joint diagonalization is computed with the frjd (fast real joint diagonalization) algorithm from the package JADE.

The available finite realizations of the space time random fields can be defined by matrices or an object of classes STFDF, STSDF or st_sftime.

Author(s)

Christoph Muehlmann, Nikolaus Piccolotto, Claudia Cappello, Sandra De Iaco, Klaus Nordhausen

Maintainer: Klaus Nordhausen [e-mail]

References

Muehlmann, C., De Iaco, S. and Nordhausen, K. (2023), Blind Recovery of Sources for Multivariate Space-Time Environmental Data. Stochastic and Environmental Research and Risk Assessment, 37, 1593–1613, <doi:10.1007/s00477-022-02348-2>.

SpaceTimeBSS documentation built on June 22, 2024, 12:16 p.m.