Forecasting with the Theta Method : Theory and Applications

Cover -- Title Page -- Copyright -- Contents -- Author Biography -- Preface -- Part I Theory, Methods and Models -- Chapter 1 The -legacy -- 1.1 The Origins… -- 1.1.1 The Quest for Causality -- 1.2 The Original Concept: THETA as in THErmosTAt -- 1.2.1 Background: A Decomposition Approach to Forecasting -- 1.2.2 The Original Basic Model of the Theta Method -- 1.2.3 How to Build and Forecast with the Basic Model -- 1.2.4 SES with Drift -- 1.2.5 The Exact Setup in the M3 Competition -- 1.2.6 Implementing the Basic Version in Microsoft Excel -- 1.2.7 The FutuRe is Written in R -- 1.A Appendix -- Chapter 2 From the From the -method to a -model -- 2.1 Stochastic and Deterministic Trends and their DGPs -- 2.2 The θ‐method Applied to the Unit Root with Drift DGP -- 2.2.1 Main Results -- 2.2.2 Alternative Trend Functions and the Original θ‐line Approach -- 2.2.3 Implementing the θ‐method under the Unit Root DGP -- 2.3 The θ‐method Applied to the Trend‐stationary DGP -- 2.3.1 Implementing the θ‐method under the Trend‐stationary DGP -- 2.3.2 Is the AR(1)‐forecast a θ‐forecast? -- Chapter 3 The Multivariate θ‐method -- 3.1 The Bivariate θ‐method for the Unit Root DGP -- 3.2 Selection of Trend Function and Extensions -- Part II Applications and Performance in Forecasting Competitions -- Chapter 4 Empirical Applications with the θ‐method -- 4.1 Setting up the Analysis -- 4.1.1 Sample Use, Evaluation Metrics, and Models/Methods Used -- 4.1.2 Data -- 4.2 Series CREDIT -- 4.3 Series UNRATE -- 4.4 Series EXPIMP -- 4.5 Series TRADE -- 4.6 Series JOBS -- 4.7 Series FINANCE -- 4.8 Summary of Empirical Findings -- Chapter 5 Applications in Health Care -- 5.1 Forecasting the Number of Dispensed Units of Branded and Generic Pharmaceuticals -- 5.2 The Data -- 5.2.1 Prescribed vs. Dispensed -- 5.2.2 The Dataset -- 5.3 Results for Branded -- 5.4 Results for Generic