Multiple prediction intervals for holt-winters forecasting procedure.

by Lawrence Chi-Ho Lee.Thesis (M.Phil.)--Chinese University of Hong Kong, 1998.Includes bibliographical references (leaves 91-97).Abstract also in Chinese.Chapter Chapter 1 --- Introduction --- p.1Chapter 1.1 --- The Importance of Forecasting --- p.1Chapter 1.2 --- Objective --- p.3Chapter Chapter 2 --- Holt-Winters Forecasting Procedure --- p.6Chapter 2.1 --- Exponential Smoothing and Holt-Winters Method --- p.6Chapter 2.2 --- Relationships Between Holt-Winters models and ARIMA Models --- p.13Chapter 2.2.1 --- A Steady Model --- p.14Chapter 2.2.2 --- A Growth Model --- p.15Chapter 2.2.3 --- The Three-Parameter Holt-Winters Model --- p.18Chapter 2.3 --- Some Practical Issues --- p.19Chapter 2.3.1 --- Normalizing the Seasonal Factors --- p.20Chapter 2.3.2 --- Choosing Starting Values --- p.20Chapter 2.3.3 --- Choosing the Smoothing Parameters --- p.22Chapter Chapter 3 --- Methods of Constructing Simultaneous Prediction Intervals --- p.24Chapter 3.1 --- Three Approximation Procedures --- p.25Chapter 3.1.1 --- Bonferroni-type Inequality --- p.26Chapter 3.1.2 --- Product-type Inequality --- p.28Chapter 3.1.3 --- Chi-square-type Inequality --- p.30Chapter 3.2 --- The 'Exact' Procedure --- p.31Chapter 3.3 --- Summary --- p.32Chapter Chapter 4 --- An Illustrative Example --- p.33Table 4.1 - 4.7 --- p.47Figure 4.1 - 4.5 --- p.55Chapter Chapter 5 --- Simulation Study --- p.60Chapter 5.1 --- Holt-Winters Forecasting Procedure for Optimal Model --- p.60Chapter 5.2 --- Holt-Winters Forecasting Procedure for Some Non-optimal Models --- p.66Chapter 5.3 --- A Comparison of Box-Jenkins Method and Holt-Winters Forecasting Procedure --- p.68Chapter 5.4 --- Conclusion --- p.74Table 5.1-5.10 --- p.75Chapter Chapter 6 --- Further Research --- p.82APPENDIXES --- p.87REFERENCES --- p.9