On Kalman filtering, posterior mode estimation and Fisher scoring in dynamic exponential family regression
DOI10.1007/BF02613597zbMath0713.62092MaRDI QIDQ750064
Ludwig Fahrmeir, Heinz Kaufmann
Publication date: 1991
Published in: Metrika (Search for Journal in Brave)
Full work available at URL: https://eudml.org/doc/176327
smoothing; posterior mode estimation; Gauss-Newton algorithm; extended Kalman filtering; Approximate error covariance matrices; block- bidiagonal matrices; categorical responses; counted responses; Dynamic exponential family regression; factorization of information matrices; Fisher scoring iterations; forward-backward recursive form; time dependent parameters
62M20: Inference from stochastic processes and prediction
62J02: General nonlinear regression
65C99: Probabilistic methods, stochastic differential equations
Related Items
Cites Work
- Unnamed Item
- Consistency and asymptotic normality of the maximum likelihood estimator in generalized linear models
- Regression models for nonstationary categorical time series: Asymptotic estimation theory
- Maximum likelihood estimation and large-sample inference for generalized linear and nonlinear regression models
- Dynamic Generalized Linear Models and Bayesian Forecasting
- Non-Gaussian State-Space Modeling of Nonstationary Time Series
- Composite Link Functions in Generalized Linear Models
- REGRESSION MODELS FOR NON‐STATIONARY CATEGORICAL TIME SERIES