Expert Adjustments of Model Forecasts
To what extent should anybody who has to make model forecasts generated from detailed
data analysis adjust their forecasts based on their own intuition? In this book, Philip
Hans Franses, one of Europe's leading econometricians, presents the notion that many
publicly available forecasts have experienced an 'expert's touch', and questions whether
this type of intervention is useful and if a lighter adjustment would be more beneficial.
Covering an extensive research area, this accessible book brings together current
theoretical insights and new empirical results to examine expert adjustment from an
econometric perspective.
The author's analysis is based on a range of real forecasts and the datasets upon which
the forecasters relied. The various motivations behind experts' modifications are
considered, and guidelines for creating more useful and reliable adjusted forecasts are
suggested. This book will appeal to academics and practitioners with an interest in
forecasting methodology. Advance praise: 'All economic and business forecasting involves
judgment to some extent, but most books focus on the application of statistical methods.
At last we have a book that directly addresses the role of judgment in forecasting and
applies a set of rigorous methods to assess its potential strengths and limitations.' Paul
Goodwin, Emeritus Professor, University of Bath Advance praise: 'This monograph is an
important substantial analysis of a crucial aspect of forecasting practice that is often
treated far too casually. Expert interventions on model-based forecasts are ubiquitous,
but rarely subject to formal analysis. This book will be valuable in practice and a
stimulus to researchers.' Derek Bunn, London Business School
Preface;
1. Introduction;
2. Optimal behavior of experts;
3. Observed behavior of experts;
4. How accurate are expert-adjusted forecasts?;
5. How can forecasts be improved?;
6. Conclusion, limitations and implications;
Data appendix; References; Index.
143 pages, Paperback