Stochastic thinning of nearest-neighbor pairwise composite likelihoods achieves 10-30x speedup over Vecchia approximation on multi-million-point spatial datasets with modest efficiency loss, concentrated on smoothness estimation.
ACM Transactions on Mathematical Software (TOMS) , volume=
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ME 1years
2026 1verdicts
ACCEPT 1representative citing papers
citing papers explorer
-
Fast Stochastic Nearest Neighbor Pairwise Composite Likelihood for Massive Spatial Datasets
Stochastic thinning of nearest-neighbor pairwise composite likelihoods achieves 10-30x speedup over Vecchia approximation on multi-million-point spatial datasets with modest efficiency loss, concentrated on smoothness estimation.