Forecasting with exponential smoothing: the state space approach

Forecasting with Exponential Smoothing
The State Space Approach
Hyndman, R.J., Koehler, A.B., Ord, J.K., Snyder, R.D.

2008, XII, 360 p., Softcover
ISBN: 978-3-540-71916-8

What is the book about?

Exponential smoothing methods have been around since the 1950s, and are still the most popular forecasting methods used in business and industry. However, a modelling framework incorporating stochastic models, likelihood calculation, prediction intervals and procedures for model selection, was not developed until relatively recently. Two key papers were Ord, Koehler and Snyder (JASA, 1997) and Hyndman, Koehler, Snyder and Grose (IJF, 2002) although there have been many others filling in some of the details.

As a result, the area of exponential smoothing has undergone a substantial revolution in the past ten years. The new "state space framework" for exponential smoothing is discussed in numerous journal articles but there has been no systematic explanation and development of the ideas. Furthermore, the notation used in the journal articles tends to change from paper to paper. Consequently, researchers and practitioners struggle to use the new models in applications.

In this book we try to bring together all of the important results in a coherent manner with consistent notation. We have written it for people wanting to apply the methods in their own area of interest as well as for researchers wanting to take the ideas in new directions.

The readership is assumed to have a statistical background at about honours level in the UK/Australian/NZ system and Masters level in the US system.

Part I: Introduction

Basic concepts
Getting started (sample chapter available)

Part II: Essentials

Linear innovations state space models
Non-linear and heteroscedastic innovations state space models
Estimation of innovations state space models
Prediction distributions and intervals
Selection of models

Part III: Further topics

Normalizing seasonal components
Models with regressor variables
Some properties of linear models
Reduced forms and relationships with ARIMA models
Linear innovations state space models with random seed states
Conventional state space models
Time series with multiple seasonal patterns (with Phillip Gould)
Non-linear models for positive data (with Muhammad Akram)
Models for count data
Vector exponential smoothing (with Ashton de Silva)

Part IV: Applications

Inventory control application
Conditional heteroscedasticity and applications in finance
Economic applications: the Beveridge-Nelson decomposition (with Chin Nam Low and Heather Anderson)

R packages

The forecast package for R implements the methods described in the book. The expsmooth package contains the data for the exercises and most of the examples in the book. Both packages are part of the forecasting bundle.

Data

The data are available in two forms: as part of the expsmooth package for R and as individual csv files. This zip file contains all csv files, or you can download individual csv files below.

freight: Annual US new freight cars, 1947-1993. Freight cars, new (excl. rebuilt), new orders, equip. mfrers. Series N0193 from the M3 competition.
usnetelec: Annual US net electricity generation (billion kwh) for 1949-2003.
utility: Hourly utility demand, mid western USA from 1 Jan 2003.
vehicles: Hourly vehicle counts on Monash Freeway, outside Melbourne in Victoria, Australia, beginning August 1995.
M3 competition: M3 competition data
visitors: Monthly Australian short-term overseas vistors. May 1985-April 2005.
canadagas: Monthly Canadian gas production, billions of cubic metres, January 1960–February 2005.
mcopper: Monthly copper prices. Copper, grade A, electrolytic wire bars/cathodes,LME,cash (£/t) Source: UNCTAD (http://stats.unctad.org/Handbook).
dji: Monthly Dow Jones index. Jan 1990 - Mar 2007.
djiclose: Monthly Dow Jones index. Closing values on first day of each month. Oct 1928 - Dec 2007.
xrates: Monthly exchange rates. Australian dollar/US dollar and Australian dollar/UK pound. Jan 2000 - May 2006.
hospital: Monthly hospital patient count for products used in medical procedures. Jan 2000 - Dec 2006.
carparts: Monthly sales car parts, Jan 1998 - Mar 2002.
msales: Monthly sales for a product with shortage indicators.
partx: Monthly sales of an automobile part.
unemp.cci: Monthly US civilian unemployment and consumer confidence, Jun 1997 - Sep 2005.
enplanements: Monthly US domestic enplanements. (millions). Jan 1979 - Jun 2002. Source: Department of Transportation, Bureau of Transportation Statistics, Air Carrier Traffic Statistic Monthly
gasprice: Monthly US retail gasoline price (the average price per gallon, in dollars) and the spot price of a barrel of West Texas Intermediate (WTI) oil in dollars as traded at Cushing, Oklahoma. These series are available from the US Energy Information Administration website http://www.eia.doe.gov.
bonds: Monthly US government bond yields (%pa) from Jan 1994 to May 2004.
ausgdp: Quarterly Australian GDP per capita 1971:1 - 1998:1.
frexport: Quarterly exports of a French company. (in thousands of francs) taken from Makridakis et al. (1998, p.162).
ukcars: Quarterly UK passenger car production (thousands of cars). 1997:1-2005:1.
usgdp: Quarterly US GDP. 1947:1 - 2006:1.
fmsales: Weekly FM sales. (Sales of a product for 62 weeks starting in early 2003.)
jewelry: Weekly jewelry sales. (Weekly sales of 314 costume jewelry items over the period week 5, 1998 to week 24, 2000.)

Authors