Evaluating The Reliability Of Macro Economic Models

It is commonly believed that macroeconomic models are not useful for policy analysis because they do not take proper account of agents' expectations. Over the last decade, mainstream macroeconomic models in the UK and elsewhere have taken on board the `Rational Expectations Revolution' by explicitly incorporating expectations of the future. In principle, one can perform the same technical exercises on a forward expectations model as on a conventional model -- and more! Rational Expectations in Macroeconomic Models deals with the numerical methods necessary to carry out policy analysis and forecasting with these models. These methods are often passed on by word of mouth or confined to obscure journals. Rational Expectations in Macroeconomic Models brings them together with applications which are interesting in their own right. There is no comparable textbook in the literature. The specific subjects include: (i) solving for model consistent expectations; (ii) the choice of terminal condition and time horizon; (iii) experimental design: i.e., the effect of temporary vs permanent, anticipated vs. unanticipated shocks; deterministic vs. stochastic, dynamic vs. static simulation; (iv) the role of exchange rate; (v) optimal control and inflation-output tradeoffs. The models used are those of the Liverpool Research Group in Macroeconomics, the London Business School and the National Institute of Economic and Social Research.

This book gives a practical, applications-oriented account of the latest techniques for estimating and analyzing large, nonlinear macroeconomic models. Ray Fair demonstrates the application of these techniques in a detailed presentation of several actual models, including his United States model, his multicountry model, Sargent's classical macroeconomic model, autoregressive and vector autoregressive models, and a small (twelve equation) linear structural model. He devotes a good deal of attention to the difficult and often neglected problem of moving from theoretical to econometric models. In addition, he provides an extensive discussion of optimal control techniques and methods for estimating and analyzing rational expectations models. A computer program that handles all the techniques in the book is available from the author, making it possible to use the techniques with little additional programming. The book presents the logic of this program. A smaller program for personal microcomputers for analysis of Fair's United States model is available from Urban Systems Research & Engineering, Inc. Anyone wanting to learn how to use large macroeconomic models, including researchers, graduate students, economic forecasters, and people in business and government both in the United States and abroad, will find this an essential guidebook.

This is the fourth version of a model that five years ago we set out to build and estimate along the lines of the continuous time approach clarified In chapter 1. Previous versions appeared in journal articles and conference proceedings, where the space is notoriously limited. Therefore we welcome the possibility of publishing a book-length treatment of this fourth version, so that we can describe its theoretical and empirical aspects in some detail. Although we have worked closely together and accept joint responsibility for the whole book, chs. 1 and 2 and appendix I have been written by G. Gandolfo, whilst chs. ] and 4 and appendix II have been written by P.c. Padoan. Different parts of this version of the model have been discussed In various lectures at the European University Institute (Florence) in 1984, In a seminar organized by the Bank of Italy (Sadiba, Perugia, Italy, February 16-18, 1984), in the second Viennese Workshop on Economic Applications of Control Theory (Vienna, May 16-18, 1984), and in the sixth annual Conference of the Society for Economic Dynamics and Control (Nice, France, June 13-15, 1984). In all of these we received helpful comments; similarly helpful were the comments of Clifford R .. Wymer, who, however, is absolved of any responsibility.

This book collects the revised and edited proceedings of the conference held in honour of the 50th anniversary of Professor Tinbergen's first macroeconomic policy model. Written by experts both in the field of model building and policy analysis, the contributions provide an invaluable overview of the state of the art and the use of macroeconomic models in our time.

Advances in computer technology, coupled with the sophistication of econometric modelling, have enabled rapid progress in the formulation and solution of optimal control and filtering programmes, especially in the sphere of macroeconomic policy designing. These developments in systems methodology have prompted the need for an interface between optimal control theory and dynamic macroeconomic analysis. The implications of this convergence have already aroused a great deal of research, but it remains to be seen whether policy makers in most developing countries will consider actually incorporating these techniques into planning. The author argues that control and systems theory can be of immense help in stabilizing those economies plagued by cyclical and structural problems. By demonstrating the applicability of control & filter theory to short-term macroeconomic planning, this book illuminates the impressive array of problems that can thereby be solved, and helps foster a closer working relationship between economists and control theorists. The work deals specifically with the construction of a Kalman filter mechanism, for deriving short-term optimal economic policies under conditions of uncertainty. It specifies and resolves a macroeconometric model which is linked to a unique observation sub-system of a given economy, congruent with the errors in information signalling which are prevalent within the data base context of most developing countries. An evaluation of control settings contrasts short and long-term economic policies. This indicates that an economy may `overheat' under protracted settings of instrument values around their optimal levels if the constraints on the system, in the form of external shocks, are too great to allow reaching all targets simultaneously using feasible instrument paths.

A Companion to Economic Forecasting provides an accessible and comprehensive account of recent developments in economic forecasting. Each of the chapters has been specially written by an expert in the field, bringing together in a single volume a range of contrasting approaches and views. Uniquely surveying forecasting in a single volume, the Companion provides a comprehensive account of the leading approaches and modeling strategies that are routinely employed.