Predicting Time-Varying Parameters with Parameter-Driven and Observation-Driven Models

By S. J. Koopman, André Lucas, Marcel Scharth. Review of Economics and Statistics, 2016

139 citations recorded by the source corpus.

How researchers see this paper

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

Time‐Varying Transition Probabilities for Markov Regime Switching Models (2017, methodology): “…model is misspecified, the score driven specification appears to be highly competitive to the correctly specified statistical model in terms of its forecasting performance; similar findings, but in the context of nonlinear non-Gaussian state space models, are also reported in Koopman et al. (2016).”
Forecasting football match results in national league competitions using score-driven time series models (2019, result): “Koopman et al. (2016) argued that the forecasting performances of univariate score-driven models are comparable to those of their parameter-driven (state space model) counterparts.”
Information Theoretic Optimality of Observation Driven Time Series Models (2014, unknown): “Harvey (2013), Koopman, Lucas, and Scharth (2012), Andres (2014), and Harvey and Luati (2014) propose different dynamic location and scale models that are specific applications of the predictive score framework, while new dynamic copula models have been investigated by Oh and Patton (2013) and De…”