A unified square-root approach for the score and Fisher information matrix computation in linear dynamic systems

DOI10.1016/j.matcom.2015.07.007OpenAlexW1655134029WikidataQ57604748 ScholiaQ57604748MaRDI QIDQ2228730

Yanyan Li

Publication date: 19 February 2021

Published in: Mathematics and Computers in Simulation (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1016/j.matcom.2015.07.007

zbMATH Keywords

maximum likelihood estimationKalman filteringFisher information matrixgradient-based methodssquare-root algorithms

Mathematics Subject Classification ID

Numerical analysis (65-XX) Combinatorics (05-XX)

Cites Work

• On efficient parametric identification methods for linear discrete stochastic systems
• Maximum likelihood estimation via the extended covariance and combined square-root filters
• Factorization methods for discrete sequential estimation
• Round-off error propagation in four generally-applicable, recursive, least-squares estimation schemes
• A general approach to constructing parameter identification algorithms in the class of square root filters with orthogonal and \(J\)-orthogonal tranformations
• Extension of square-root filtering to include process noise
• Constructing numerically stable Kalman filter-based algorithms for gradient-based adaptive filtering
• Reduced gradient computation in prediction error identification
• Numerical aspects of different Kalman filter implementations
• A new method for evaluating the log-likelihood gradient (score) of linear dynamic systems
• A new method for evaluating the log-likelihood gradient, the Hessian, and the Fisher information matrix for linear dynamic systems
• Derivative computations for the log likelihood function
• Gradient computation in prediction error identification of linear discrete-time systems
• Computational aspects of maximum likelihood estimation and reduction in sensitivity function calculations
• Square-root algorithms for least-squares estimation
• Efficient computation of gradient and hessian of likelihood function in linear dynamic systems
• Some new algorithms for recursive estimation in constant linear systems
• Extended Chandrasekhar recursions
• J-Orthogonal Matrices: Properties and Generation
• Array algorithms for H/sup ∞/ estimation
• New square-root algorithms for Kalman filtering
• Likelihood Gradient Evaluation Using Square-Root Covariance Filters
• Maximum likelihood estimation using square root information filters
• Kalman Filter Sensitivity Evaluation With Orthogonal and J-Orthogonal Transformations
• State Sensitivity Evaluation Within UD Based Array Covariance Filters
• Approaches to adaptive filtering

This page was built for publication: A unified square-root approach for the score and Fisher information matrix computation in linear dynamic systems