A validation-driven clustering framework groups multivariate time series by predictive performance for adaptive pooling, with a leakage-free fallback to global models when specialization does not help.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ME 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Forecasting Multivariate Time Series under Predictive Heterogeneity: A Validation-Driven Clustering Framework
A validation-driven clustering framework groups multivariate time series by predictive performance for adaptive pooling, with a leakage-free fallback to global models when specialization does not help.