This book is the first to detail the Theta method of forecasting. The Theta method of Forecasting is considered one of the most difficult-to-beat forecasting benchmarks and has topped the biggest ever forecasting competition in the world - the M3 competition, which in 2000, involved 19 academic methods and six software packages to compete for what would provide the best forecasts for 3003 real series from industry, finance and economics.
Written by the two leading experts in the field, this book will throw light on the exact replication of the method and under what conditions the method outperforms all other forecasting methods. Recent developments such as a multivariate step are also included and as such this book is a valuable tool for both academics and practitioners
The book includes three main parts:
Part I: Method, Models and Applications - detailing new theory about the model
Part II: Performance in Forecasting Competitions - detailing empirical result and simulations on the method
Part III: Contributions - contributions from other academics working on the method in the recent years