Maximum likelihood estimation for score-driven models

By Francisco Blasques, Janneke van Brummelen, S. J. Koopman, André Lucas. Journal of Econometrics, 2021

80 citations recorded by the source corpus.

Abstract We establish strong consistency and asymptotic normality of the maximum likelihood estimator for stochastic time-varying parameter models driven by the score of the predictive conditional likelihood function. For this purpose, we formulate primitive conditions for global identification, invertibility, strong consistency, and asymptotic normality both under correct specification and misspecification of the model. A detailed illustration is provided for a conditional volatility model with disturbances from the Student’s t distribution.

Subfields: Economics and Econometrics, Finance

View Paper: https://www.semanticscholar.org/paper/9c3ba6972c0a89bf80b68177a5a18b36c9c2ee4c

Citations in the corpus

• Value-at-risk and expected shortfall in cryptocurrencies’ portfolio: a vine copula–based approach (2019, unknown): “This choice is particularly attractive for its flexibility to capture the volatility dynamics and for its robustness to heavy tails and outliers – for example, see Harvey and Chakravarty (2008), Harvey and Sucarrat (2014), Blazsek and Villatoro (2015) and Blasques, Koopman, and Lucas (2017).”