Computational Statistics

Research Areas

• Statistical analysis of time series and signal processing: algorithms for identification, estimation and forecasting methods.
• Models with time-dependent coefficients, non-linear models (threshold models, ARCH models), intervention models.
• Development of an Expert System for Economic Forecasting: TSE (Time Series Expert).
• Empirical macroeconomics and dynamics of large cross-sectional data.

Selected Publications

• Azrak, R., G. Mélard (1998). The exact quasi-likelihood of time-dependent ARMA models, Journal of Statistical Planning and Inference, 68, 31-45.
• G. Mélard (1990). Méthodes de prévision à court terme , Coll. Statistique et Math?matiques appliquées, Editions de l'Université de Bruxelles, Bruxelles et Editions Ellipses, Paris.
• Klein, A., G. Mélard and T. Zahaf (1998). Computation of the exact information matrix of Gaussian dynamic regression time series models, Annals of Statistics, 26, 1636-1650.

Members

Contact