Time Series Analysis of Non-Gaussian Observations Based on State Space Models from Both Classical and Bayesian Perspectives

By J. Durbin, S. J. Koopman. 1999

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Effective Bayesian Modeling of Groups of Related Count Time Series (2014, background): “Durbin & Koopman (2000) present an alternative approach based on importance sampling and antithetic variables.”
Estimating a State-Space Model from Point Process Observations (2003, methodology): “The Fahrmeir forward recursion algorithm for example 1 with a single neuron is whereas the Durbin and Koopman (2000) update is The variance updating algorithm in the Fahrmeir algorithm agrees because the observed and expected Fisher information are the same for the Poisson model in our example.”
Bootstrap Approximation to Prediction MSE for State–Space Models with Estimated Parameters (2005, methodology): “The need of developing valid PMSE estimators for state–space models with estimated parameters is often raised by researchers working in this field (see, e.g. Durbin and Koopman, 2000 and the discussion of A. Harvey to that article).”
Monte Carlo methods for estimating, smoothing, and filtering one- and two-factor stochastic volatility models (2006, methodology): “This paper demonstrates tools for likelihood-based analysis of standard one- and two-factor SV models used in finance, building on work by Kim et al. (1998), Shephard and Pitt (1997), Durbin and Koopman (1997, 2000), and Sandmann and Koopman (1998).”
Double hierarchical generalized linear models (with discussion) (2006, methodology): “…936 mean-corrected returns yt =100 { log.rt=rt−1/− ∑ log.ri=ri−1/ n } , where rt denotes the exchange rate at time t. Harvey et al. (1994), Shephard and Pitt (1997), Kim et al. (1998) and Durbin and Koopman (2000) fitted the SV model log.φt/=γ0 +bt , .8/ where bt =ρbt−1 + rt ∼AR.1/ with rt ∼N.0,α/.”
Estimation methods for stochastic volatility models: a survey (2004, methodology): “If the posterior mean is required, it is possible to use the following computationally efficient algorithm of Durbin and Koopman (2000): log 2t=T ¼ log y2t XN i¼1 ft e (i)t # Blackwell Publishing Ltd. 2004 where ft ¼ !ð̂ (i) t Þ N!”
Estimating common trends in multivariate time series using dynamic factor analysis (2003, methodology): “Fahrmeir and Tutz (1994) and Durbin and Koopman (2000) used state-space models in which the errors terms were Poisson and Binomial distributed.”
The structure of dynamic correlations in multivariate stochastic volatility models (2009, methodology): “With regard to block samplers, there are various block samplers, including Durbin and Koopman (2000), Garmerman (1998) and Shephard and Pitt (1997).”
Predicting Time-Varying Parameters with Parameter-Driven and Observation-Driven Models (2016, background): “…the stochastic conditional duration model of Bauwens and Veredas (2004); the stochastic conditional intensity model of Bauwens and Hautsch (2006); the stochastic copula models of Hafner and Manner (2012); and the non-Gaussian unobserved components time series models of Durbin and Koopman (2000).”
Double hierarchical generalized linear models (2006, methodology): “The state space models of Harvey (1989) and Durbin and Koopman (2000) and the spatial models of Besag and Higdon (1999) are models in which L.ρ/=L, i.e. there are no unknown parameters.”
Time Series and Forecasting: Brief History and Future Research (2000, background): “Durbin and Koopman (2000) provided both classical and Bayesian perspectives in time series analysis.”
Monte Carlo State-Space Likelihoods by Weighted Posterior Kernel Density Estimation (2004, background): “…sampling (Geyer and Thompson 1992; Geyer 1994, 1996; McCulloch 1997); basic MC integration;more advanced MC integration,such as importance sampling (Durbin and Koopman 1997, 2000); and particle ltering (PF) (Gordon, Salmond, and Smith 1993;Kitagawa 1996, 1998; Pitt and Shephard 1999; Doucet, de…”
Bayesian predictions of low count time series (2005, methodology): “We refer to the binomial thinning process in (10), combined with negative binomial arrivals, as the negative binomial autoregressive (NBAR) model, with this model being characterized by the three-dimensional parameter vector θ3 = (α, π, r)0.”
Energy markets and CO2 emissions: Analysis by stochastic copula autoregressive model (2015, unknown): “A 7 particular case of the latter is the method of importance sampling, discussed by Durbin and Koopman (2000, 2001)[24] and used to estimate time series models with a state space representation.”