Results 131 to 140 of about 109,235 (253)

The Mathematical History Behind the Granger–Johansen Representation Theorem

Oxford Bulletin of Economics and Statistics, EarlyView.
ABSTRACT When can a vector time series that is integrated once (i.e., becomes stationary after taking first differences) be described in error correction form? The answer to this is provided by the Granger–Johansen representation theorem. From a mathematical point of view, the theorem can be viewed as essentially a statement concerning the geometry of ...
Johannes M. Schumacher
wiley   +1 more source

Large‐Dimensional Cointegrated Threshold Factor Models: The Global Term Structure of Interest Rates

Oxford Bulletin of Economics and Statistics, EarlyView.
ABSTRACT In this paper we extend the two‐level factor model to account for cointegration between group‐specific factors in large datasets. We propose two nonlinear specifications: (i) a threshold vector error correction model (VECM) that allows for asymmetric adjustment across regimes; and (ii) a band VECM that captures state‐dependent adjustment which
Daniel Abreu, Paulo M. M. Rodrigues
wiley   +1 more source

Repelled Point Processes With Application to Numerical Integration

Scandinavian Journal of Statistics, EarlyView.
ABSTRACT We look at Monte Carlo numerical integration from a stochastic geometry point of view. While crude Monte Carlo estimators relate to linear statistics of a homogeneous Poisson point process (PPP), linear statistics of more regularly spread point processes can yield unbiased estimators with faster‐decaying variance, and thus lower integration ...
Diala Hawat, Gabriel Mastrilli, Rémi Bardenet, Raphaël Lachièze‐Rey   +3 more
wiley   +1 more source

Sparse Minimum Redundancy Maximum Relevance for Feature Selection

Scandinavian Journal of Statistics, EarlyView.
ABSTRACT We propose a feature screening method that integrates both feature–feature and feature–target relationships. Inactive features are identified via a penalized minimum Redundancy Maximum Relevance (mRMR) procedure, which is the continuous version of the classical mRMR penalized by a non‐convex regularizer, and where the parameters estimated as ...
Peter Naylor, Benjamin Poignard, Héctor Climente‐González, Makoto Yamada   +3 more
wiley   +1 more source

Compositional solution of stochastic process algebra models.

, 2002
This dissertation is about the solution of Markovian stochastic process algebra (SPA) models and the avoidance of the state-space explosion problem. We try to answer the question whether the compositionality of SPA models can be exploited to overcome the largeness problems appearing when evaluating such models.
openaire   +2 more sources

Sequential Monte Carlo with likelihood tempering and parallel implementation for uncertainty quantification

AIChE Journal, Volume 72, Issue 6, June 2026.
Abstract Bayesian estimation enables uncertainty quantification, but analytical implementation is often intractable. As an approximate approach, the Markov Chain Monte Carlo (MCMC) method is widely used, though it entails a high computational cost due to frequent evaluations of the likelihood function.
Tatsuki Maruchi, Tomoyuki Yajima, Yoshiaki Kawajiri   +2 more
wiley   +1 more source

Ramifications of generalized Feller theory. [PDF]

J Evol Equ
Cuchiero C, Möllmann T, Teichmann J.
europepmc   +1 more source

Topological Graph Neural Networks: A Novel Approach for Geometric Deep Learning

Applied AI Letters, Volume 7, Issue 2, June 2026.
This graphical abstract illustrates the Topological Graph Neural Network (TopGNN) architecture. It demonstrates a parallel processing approach where an input graph is simultaneously analyzed by a standard GNN Encoder to capture local node features and by Persistent Homology to extract global topological features (like cycles and voids), visualized as a
Amarjeet, Anurag Sinha, Khalid Hussain, Wasim Ahmad Bhat, Sudhanshu Maurya, Mohammad Nadeem Ahmed, Mohammad Rashid Hussain, S. Khalid   +7 more
wiley   +1 more source

Thermodynamics à la Souriau on Kähler Non-Compact Symmetric Spaces for Cartan Neural Networks. [PDF]

Entropy (Basel)
Fré PG, Sorin AS, Trigiante M.
europepmc   +1 more source

Fourier Mass Lower Bounds for Batchelor‐Regime Passive Scalars

Communications on Pure and Applied Mathematics, Volume 79, Issue 6, Page 1449-1466, June 2026.
ABSTRACT Batchelor predicted that a passive scalar ψν$\psi ^\nu$ with diffusivity ν$\nu$, advected by a smooth fluid velocity, should typically have Fourier mass distributed as |ψ̂ν|2(k)≈|k|−d$|\widehat{\psi }^\nu |^2(k) \approx |k|^{-d}$ for |k|≪ν−1/2$|k| \ll \nu ^{-1/2}$.
William Cooperman, Keefer Rowan
wiley   +1 more source