Forecasting: Principles and Practice (Hyndman)

Based on R

Hyndman, R.J. and Athanasopoulos, G. (2013) Forecasting: principles and practice. http://otexts.org/fpp/. Accessed on January 2014

From the book's website:

• "Welcome to our new online textbook on forecasting. This book is a replacement for Makridakis, Wheelwright and Hyndman (Wiley1998)"

Table of Contents

Forecasting: principles and practice

Getting started

• What can be forecast?

• Forecasting, planning and goals

• Determining what to forecast

• Forecasting data and methods

• Some case studies

• The basic steps in a forecasting task

• The statistical forecasting perspective

• Exercises

• Further reading

The forecaster's toolbox

• Graphics

• Numerical data summaries

• Some simple forecasting methods

• Transformations and adjustments

• Evaluating forecast accuracy

• Residual diagnostics

• Prediction intervals

• Exercises

• Further reading

• The forecast package in R

Judgmental forecasts

• Introduction

• Beware of limitations

• Key principles

• The Delphi method

• Forecasting by analogy

• Scenario forecasting

• New product forecasting

• Judgmental adjustments

• Further reading

Simple regression

• The simple linear model

• Least squares estimation

• Regression and correlation

• Evaluating the regression model

• Forecasting with regression

• Statistical inference

• Non-linear functional forms

• Regression with time series data

• Summary of notation and terminology

• Exercises

• Further reading

Multiple regression

• Introduction to multiple linear regression

• Some useful predictors

• Selecting predictors

• Residual diagnostics

• Matrix formulation

• Non-linear regression

• Correlation, causation and forecasting

• Exercises

• Further reading

Time series decomposition

• Time series components

• Moving averages

• Classical decomposition

• X-12-ARIMA decomposition

• STL decomposition

• Forecasting with decomposition

• Exercises

• Further reading

Exponential smoothing

• Simple exponential smoothing

• Holt's linear trend method

• Exponential trend method

• Damped trend methods

• Holt-Winters seasonal method

• A taxonomy of exponential smoothing methods

• Innovations state space models for exponential smoothing

• Exercises

• Further reading

ARIMA models

• Stationarity and differencing

• Backshift notation

• Autoregressive models

• Moving average models

• Non-seasonal ARIMA models

• Estimation and order selection

• ARIMA modelling in R

• Forecasting

• Seasonal ARIMA models

• ARIMA vs ETS

• Exercises

• Further reading

Advanced forecasting methods

• Dynamic regression models

• Vector autoregressions

• Neural network models

• Forecasting hierarchical or grouped time series

• Further reading

Data

• All data sets from the book are provided in the fpp package for R.

Using R

• Installing R and RStudio

• R examples in this book

• Getting started with R

• More tutorials

Resources

• Here is a one-hour talk by one of the authors on Forecasting time series using R.

• Slides

1. Introduction to forecasting [Exercises, Solutions]

2. The forecaster's toolbox [Exercises, Solutions]

3. Autocorrelation and seasonality [Exercises, Solutions]

4. White noise and time series decomposition [Exercises, Solutions]

5. Exponential smoothing methods [Exercises, Solutions]

6. ETS models [Exercises, Solutions]

7. Transformations and adjustments [Exercises, Solutions]

8. Stationarity and differencing [Exercises, Solutions]

9. Non-seasonal ARIMA models [Exercises, Solutions]

10. Seasonal ARIMA models [Exercises, Solutions]

11. Dynamic regression [Exercises, Solutions]

12. Advanced methods

Reviews