Time Series Analysis by State Space Methods

By J. Durbin, S. J. Koopman, S. J. Koopman. 2001

3,294 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Statistics and Probability, Time-Series & Forecasting

How researchers see this paper

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

In Artificial Intelligence

In Statistics and Probability

In Time-Series & Forecasting

Citations in the corpus

Effective Bayesian Modeling of Groups of Related Count Time Series (2014, methodology): “For a single non-negative time series, the model is expressed in state-space form (e.g., Durbin & Koopman, 2012), where the latent state ηt at period t = 1, . . . , T represents the log-expected value of the non-negative integer observation yt.”
Enhancing the Locality and Breaking the Memory Bottleneck of Transformer on Time Series Forecasting (2019, methodology): “Although still widely used, traditional time series forecasting models, such as State Space Models (SSMs) [2] and Autoregressive (AR) models, are designed to fit each time series independently.”
Automatic Time Series Forecasting: The forecast Package for R (2008, methodology): “One solution to this diﬃculty is the “diﬀuse prior’ ’ approach which is outlined in Durbin and Koopman (2001) and implemented in the arima() function (Ripley 2002) in R .”
Fixed-Form Variational Posterior Approximation through Stochastic Linear Regression (2012, methodology): “The approximate marginal likelihood q η ( y | φ, σ 2 ) and the expectations with respect to q η ( f | φ, σ 2 ) can be evaluated analytically using the Kalman ﬁlter and smoother (e.g. Durbin and Koopman, 2001), which means we do not have to sample f for this problem.”
A note on composite likelihood inference and model selection (2005, methodology): “A known solution to this problem involves computational intensive, simulation-based, approximations (see, for example, Durbin & Koopman (2001) and references therein).”
Cointegration Analysis with Mixed-Frequency Data (2007, methodology): “In Appendix B, we adopt the fixedinterval smoothing algorithm of De Jong (1989), which avoids inversion of large matrices, and, hence, is computationally more efficient than the classical smoothing equations (Durbin and Koopman, 2001).”
Combining day-ahead forecasts for British electricity prices (2013, methodology): “For this model parameters were estimated using state space methods and the Kalman filter (Durbin and Koopman, 2001; Hamilton, 1994).”
Statistical inference for dynamical systems: A review (2012, methodology): “Furthermore, let us mention the following books or monographs related to the topics in this survey: [1, 8, 14, 29, 62, 64, 91, 118].”
Stochastic model specification search for Gaussian and partial non-Gaussian state space models (2010, methodology): “In step (b), forward-filtering-backward-sampling (FFBS, Frühwirth-Schnatter (1994); Carter and Kohn (1994); De Jong and Shephard (1995)) is used to sample x = (x1, .”
Predicting the present with Bayesian structural time series (2014, background): “Readers interested in the computational details of the Kalman ﬁlter and smoother can consult Durbin and Koopman (2001) or numerous other sources.”
The Analysis of Stochastic Volatility in the Presence of Daily Realised Measures (2011, methodology): “: +31(0)20 525 8579 The Analysis of Stochastic Volatility in the Presence of Daily Realised Measures∗ Siem Jan Koopman(a,b) Marcel Scharth(a,b) (a) VU University Amsterdam, The Netherlands (b) Tinbergen Institute, The Netherlands This version: September 2011 Abstract We develop a systematic framewor”
Gaussian variational approximation with sparse precision matrices (2016, methodology): “This dataset has been considered by Harvey et al. (1994), Kim et al. (1998) and Durbin and Koopman (2012).”
Modelling and forecasting fossil fuels, CO2 and electricity prices and their volatilities (2013, background): “, r, (16) where It−1 refers to all the available information at time t − 1.”