Asymptotic bias for quasi-maximum-likelihood estimators in conditional heteroskedasticity models

Article Abstract:

Almost all studies presume that time-varying conditional variance, also referred to as conditional heteroskedasticity, utilizes a quasi-maximum-likelihood estimator (QMLE). QMLE consistency assumes an identification condition that the quasi-log-likelihood have a distinctive maximum at the true conditional mean and relative scale parameters. It is concluded that the identification condition supports a non-Gaussian QMLE when the conditional mean is zero or the assumed and true innovation densities are symmetric around zero.

Models, Usage, Maximum likelihood estimates (Statistics), Gaussian processes, Maximum likelihood (Statistics)

---

Standard state-space models preclude unawareness

Article Abstract:

The role of state-space models was analyzed in relation to the existence of unawareness. It was established that the combination of the real-states and event-sufficiency assumptions contributes mainly to state-space models' common association with unawareness. The model suggests that an agent's states tend to be less complete than the analyst's whenever the agent is unaware of economic possibilities.

Economics, Research and Development in the Social Sciences and Humanities, Analysis, Axioms, Awareness

---

Similar abstracts:

• Abstracts: A semiparametric maximum likelihood estimator. Subject evaluation in social experiments. Information acquisition in auctions
• Abstracts: Maximum-likelihood estimation of fractional cointegration with an application to U.S. and Canadian bond rates
• Abstracts: Covariance estimation for multivariate conditionally Gaussian dynamic linear models. Evaluating the forecast densities of linear and non-linear models: applications to output growth and unemployment
• Abstracts: Finite sample behavior of tests for grouped heteroskedasticity. Finding cointegration rank in high dimensional systems using the Johansen test: an illustration using data based Monte Carlo simulations
• Abstracts: Asymptotic properties of equilibrium forecasts in Bayesian learning models. Observable implications of equilibrium behavior on finite data

This website is not affiliated with document authors or copyright owners. This page is provided for informational purposes only. Unintentional errors are possible.