##
**Forecasting economic time series. 2nd ed.**
*(English)*
Zbl 0642.90001

Economic Theory, Econometrics, and Mathematical Economics. Orlando etc.: Academic Press, Inc. (Harcourt Brace Jovanovich, Publishers). XIV, 338 p. (1986).

The book aims at bridging the gap between theoretical and applied aspects of forecasting. It concentrates on forecasting of economic data and therefore it combines in a balanced way typical time series procedures (e.g. Box-Jenkins methodology) with the more classical econometric approach (e.g. forecasting from regression models used in econometrics). Moreover, the combinations of various types of forecasts are emphasized as successful for real data forecasting. Some numerical examples demonstrate the described procedures. The second edition of the monograph reflects new developments in forecasting theory and practice: it expands the part devoted to multiple series modelling procedures and surveys briefly some more recent topics (e.g. state-space representation and the Kalman filter, time-varying parameter models, nonlinear models, ARCH models).

The content of the book is as follows: 1. Introduction to the theory of time series; 2. Spectral analysis (filters, spectrum and cross spectrum, estimation of spectral functions, typical spectral shape, seasonal adjustment; 3. Building linear time series models (Box-Jenkins methodology); 4. The theory of forecasting (basic concepts and properties, generalized cost function); 5. Practical methods for univariate time series forecasting; 6. Forecasting from regression models; 7. Multiple series modelling and forecasting (basic models and properties, causality and feedback, co-integrated and error-correction models); 8. Building multiple time series forecasting models (multivariate Box-Jenkins methodology, testing for causality and co- integration); 9. The combination and evaluation of forecasts; 10. Further topics (see above).

The content of the book is as follows: 1. Introduction to the theory of time series; 2. Spectral analysis (filters, spectrum and cross spectrum, estimation of spectral functions, typical spectral shape, seasonal adjustment; 3. Building linear time series models (Box-Jenkins methodology); 4. The theory of forecasting (basic concepts and properties, generalized cost function); 5. Practical methods for univariate time series forecasting; 6. Forecasting from regression models; 7. Multiple series modelling and forecasting (basic models and properties, causality and feedback, co-integrated and error-correction models); 8. Building multiple time series forecasting models (multivariate Box-Jenkins methodology, testing for causality and co- integration); 9. The combination and evaluation of forecasts; 10. Further topics (see above).

Reviewer: T.Cipra

### MSC:

91B84 | Economic time series analysis |

62P20 | Applications of statistics to economics |

62M10 | Time series, auto-correlation, regression, etc. in statistics (GARCH) |

91-01 | Introductory exposition (textbooks, tutorial papers, etc.) pertaining to game theory, economics, and finance |

62-01 | Introductory exposition (textbooks, tutorial papers, etc.) pertaining to statistics |