Modelling Economic Series: Readings in Econometric Methodology

Modelling Economic Series: Readings in Econometric Methodology

Modelling Economic Series: Readings in Econometric Methodology

Modelling Economic Series: Readings in Econometric Methodology

Synopsis

This book helps economists with the difficult task of constructing econometric models and will be especially useful to those taking courses in applied econometrics who need to learn how to evaluate the validity of the theories and techniques they are taught. The volume contains seventeenpapers by the leading authorities in the field, divided into four groups, to each of which the editor provides an introduction. The whole volume is prefaced with an editorial discussion of the controversies of the subject. The methods critically discussed include the traditional ones, such as vector auto-regressions; Bayesian techniques; and the comprehensive modelling strategy advocated by reseachers at the London School of Economics. The papers vary in the degree of sophistication used, but anyone reading the book should gain a sound knowledge of the practical difficulties involved in model specification, evaluation, and interpretation.

Excerpt

C. W. J. Granger

The novice empirical modeller faces many difficulties and problems concerned with the evolving specification, evaluation, and interpretation of the model. the advice that econometric modelling is an activity that should not be attempted for the first time is unhelpful but contains a useful warning. Good modelling is probably both an art and a science. a sound knowledge of theoretical econometrics and the availability of an efficient and flexible computer program is not enough to ensure success.

The purpose of this book is to bring together advice and discussions by prominent thinkers in the area of econometric methodology. There is certainly no one generally acceptable path to a good model. in fact, the papers included here are partially chosen to illustrate the deep differences of opinion between various groups of econometricians, concerning the modelling process and the interpretation of models. One ends with several different modelling strategies, each with supporters and critics. At the present stage of development in economics it is probably an advantage to have different groups look at the same problem from different viewpoints, so that their models and conclusions can be compared and possibly then form the basis for a new, comprehensive model. the fact that one model dominates another under some criterion, such as forecasting ability, does not necessarily imply that the better model should be retained and the lesser one totally discarded, as both models may contain elements of 'truth' and so should be combined.

The book is concerned with the process of modelling a group of economic time series for some specific purpose. It will concentrate on the strategy of modelling, the choice of specification and its evaluation when alternatives are considered, and also the evaluation of the model. Very little attention is paid to the question of how best to estimate the parameters of the model, a question to which econometricians have paid an undue amount of attention in the past 50 years.

The modelling exercise will have a dependent variable y , which may be a vector, to be explained by the contents of an information set I . This . . .

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.