No packages match

fastLISA - Fast Local Indicators of Spatial Association (LISA)

Computes various Local Indicators of Spatial Association (LISA) statistics, including univariate and bivariate local Moran's I, Empirical Bayes local Moran's I, univariate and multivariate local Geary's C, and Getis-Ord G and G* statistics. The methods follow Anselin (1995), Getis and Ord (1992), and Anselin (2019). Leverages a high-performance, plain-C backend with optional 'OpenMP' multi-core support for fast permutation-based pseudo-p-value calculation. Accepts any 'spdep' listw spatial weight matrix, including custom and non-contiguity weights. Uses sample standardisation (n-1) and 'rgeoda'-style permutation p-values. Output cluster codes match 'rgeoda' conventions, including the Isolated category for observations without neighbours.

Last updated

openmp

4.48 score 2 stars 8 scripts

IBGS - Iterated Block Gibbs Sampler for Ultrahigh-Dimensional Variable Selection and Model Averaging

Variable selection for generalized linear models and the Cox proportional-hazards model in ultrahigh dimensions via the iterated block Gibbs sampler (IBGS). The sampler is implemented in C with parallel block screening through 'OpenMP', and supports the gaussian, binomial and poisson families (fitted by least squares or iteratively reweighted least squares) as well as the Cox model for survival analysis (fitted by its Efron partial likelihood), together with the AIC, BIC, AICc and extended BIC model selection criteria.

Last updated

gibbsmethodphdopenblasopenmp

4.32 score 1 stars 14 scripts