Results 171 to 180 of about 307,922 (236)

A structurally localized ensemble Kalman filtering approach

open access: yesQuarterly Journal of the Royal Meteorological Society, EarlyView.
We derive an inherently localized ensemble Kalman filtering (EnKF) approach, avoiding the need for any auxiliary localization technique. The idea is to first use the variational Bayesian optimization to approximate the (continuous) state analysis probability density function (pdf) by a product of independent marginal pdfs corresponding to small ...
Boujemaa Ait‐El‐Fquih   +1 more
wiley   +1 more source

Hybrid physics–data‐driven modeling for sea ice thermodynamics and transfer learning

open access: yesQuarterly Journal of the Royal Meteorological Society, EarlyView.
Icepack–NN, a machine‐learning‐based hybrid version of the sea‐ice column model Icepack, is developed to correct state‐dependent forecast errors arising from misspecified snow thermodynamics, using neural networks applied online within the physical model.
G. De Cillis   +7 more
wiley   +1 more source

Adaptive CUSUM Chart for Simultaneous Monitoring of Mean and Variance

open access: yesQuality and Reliability Engineering International, EarlyView.
ABSTRACT Simultaneously monitoring changes in both the mean and variance is a fundamental problem in statistical process control, and numerous methods have been developed to address it. However, many existing approaches face notable limitations: Some rely on tuning parameters that can significantly affect performance; others are biased toward detecting
Gokul Parakulum, Jun Li
wiley   +1 more source

Inference on function-valued parameters using a restricted score test. [PDF]

open access: yesJ R Stat Soc Series B Stat Methodol
Hudson A, Carone M, Shojaie A.
europepmc   +1 more source

A Robust Self‐Starting Bayesian Approach for Multivariate Phase II Monitoring

open access: yesQuality and Reliability Engineering International, EarlyView.
ABSTRACT Traditional multivariate control charts require in‐control (IC) parameter estimates to be known or estimated from a large set of uncontaminated, historical Phase I observations. However, some processes need to be monitored when little Phase I data are available, and self‐starting approaches, including Bayesian methods, have proven useful. Self‐
Taylor R. Grimm   +2 more
wiley   +1 more source

Home - About - Disclaimer - Privacy