Quasi Score-Driven Models
The paper “Quasi score-driven models”, co-authored by Sébastien Laurent, a co-developer of OxMetrics, introduces the new class of quasi score-driven (QSD) models.
This new class inherits and extends the basic ideas behind the development of SD models and addresses several unsolved issues in the score literature. In particular, the new class of models:
- Generalizes many existing models, including SD models
- Disconnects the updating equation from the log-likelihood implied by the conditional density of the observations
- Allows testing of the assumptions behind SD models that link the updating equation of the conditional moment to the conditional density
- Allows QML estimation of SD models
- Allows explanatory variables to enter the updating equation.
The authors establish the asymptotic properties of the QLE, QMLE and MLE of the proposed QSD model, as well as the likelihood ratio and Lagrange multiplier test statistics. The finite sample properties are studied by means of an extensive Monte Carlo study. Finally, they show the empirical relevance of QSD models to estimate the conditional variance of 400 US stocks.
This article originated from the paper written by F. Blasques, Christian Francq and Sébastien Laurent, which can be read here, Quasi score-driven models.
|