Continuous updating gmm estimator

by  |  09-Jan-2015 12:55

- Finite sample properties of an estimator - Large sample properties of an estimator - Almost sure convergence - Convergence in probability and law of large numbers - Convergence in mean square - Convergence in distribution and Central Limit Theorem - Asymptotic distribution - Continous mapping theorem and delat method (slides) and (Matlab Codes) - Introduction - Likelihood function - Maximum Likelihood Estimator - Score, Gradient, Hessian and Fisher information matrix - Asymptotic properties of the maximum likelihood estimator - Application to the multiple linear regression model - Application to the probit and logit model (slides) and (Matlab Codes) - Introduction - The multiple linear regression model - Parametric and semi-parametric specifications - The Ordinary Least Squares (OLS) estimator - Statistical properties of the OLS - Finite sample properties of the OLS - Asymptotic properties of the OLS (slides) (videos) and (Matlab Codes) - Introduction - Statistical hypothesis testing and inference - Tests in the multiple linear regression model - The Student t-test - The Fisher test - Maximum Likelihood Estimation (MLE) and Inference - The Likelihood Ratio (LR) test - The Wald test - The Lagrange Multiplier (LM) or score test (slides) and (Matlab Codes) - The generalized linear regression model - Inefficiency of the Ordinary Least Squares - Generalized Least Squares (GLS) - Feasible Generalized Least Squares (FGLS) - Heteroscedasticity - White correction for heteroscedasticity - OLS and robust inference - Testing for heteroscedasticity: Breusch-Pagan and White tests (statement) and (Correction)- MLE and Weibull distribution- Wald test, LM test, LR test- OLS and multiple linear regression model __________________________ Site Value-at-Risk : Prvisions de Value-at-Risk et Backtesting Consultez le site Value-at-Risk ddi aux prvisions de Value-at-Risk (modles GARCH univaris et mthodes non paramtriques) et aux procdures de Backtesting : Estimation (Maximum de Vraisemblance et Pseudo Maximum de Vraisemblance) - Distributions Conditionnelles des modles GARCH (Student, Skewed Student et GED) - Tests d'effets ARCH - Modles GARCH asymtriques (EGARCH, QGARCH, LSTGARCH, ANSTGARCH, TGARCH, GJR-GARCH..) - Applications sous SAS : model GARCH et Value-at-Risk- Introduction : qu'est ce que le backtesting ?- Dfinitions : violation de la Va R, couverture non conditionnelle et conditionnelle.

Chapitre 3 (pdf) : Modles Variable Dpendante Limite ou Censure : Modle Tobit Simple, Modles Tobit Gnraliss.

Annexe (pdf) : Estimation Non Paramtrique et Semi Paramtrique d’un Modle Dichotomique (Klein et Spady, 1993).

Mai 2005 – Examen (nonc) Modle Logit conditionnel (Mc Fadde, 1976), modle Tobit. Tokpavi, nonc) Modle Tobit simple, mthode d’estimation d’Heckman en deux tapes, modle Logit multinomial.

Mai 2006 – Examen (nonc, correction) , Rating, modle logit multinomial ordonn, modle Tobit.

Mai 2007 – Examen (nonc, donnes format Eviews ) , Scoring sur dfaillance d'entreprises, rgression logisitique et modle semi-paramtrique, rating et modle Probit multinomial ordonn, modle Tobit. Tokpavi, nonc) Modles Probit et Logit, effets marginaux, estimation par maximum de vraisemblance.

Community Discussion