218 pages - Publisher: Springer; 2nd ed. 2009 edition (December 12, 2008)
Language: English - ISBN-10: 3540859950 - ISBN-13: 978-3540859956
This second edition sees the light three years after the first one: too short a time to feel seriously concerned to redesign the entire book, but sufficient to be challenged by the prospect of sharpening our investigation on the working of econometric dynamic models and to be inclined to change the title of the new edition by dropping the “Topics in” of the former edition. After considerable soul searching we agreed to include several results related to topics already covered, as well as additional sections devoted to new and sophisticated techniques, which hinge mostly on the latest research work on linear matrix polynomials by the second author. This explains the growth of chapter one and the deeper insight into representation theorems in the last chapter of the book. The rôle of the second chapter is that of providing a bridge between the mathematical techniques in the backstage and the econometric profiles in the forefront of dynamic modelling. For this purpose, we decided to add a new section where the reader can find the stochastic rationale of vector autoregressive specifications in econometrics. The third (and last) chapter improves on that of the first edition by re- ing the fruits of the thorough analytic equipment previously drawn up.
From the Back Cover: This monograph provides an insightful analysis of dynamic modelling in econometrics by bridging the structural with the time series approaches, and by focusing on representation theorems of integrated processes. The book provides mainly a self-contained, rigorous as well as innovative, analytic setting to guide formulation and solution in closed form of vector autoregressive (VAR) models with unit roots. The second edition implements the latest research work by the second author on linear matrix polynomials whence a further breakthought on the topic is gained. Its emphasis is placed on representation theorems, conjugating an elegant reappraisal of classical results with original insights which widen their information content. A unified representation theorem of new conception is established, which duly shapes the contours of the cointegration features of VAR solutions, providing not only a contribution to clarity but also new stimuli in this fascinating field of research as a spin-off.