на главную поиск contacts

Time Series Models

Опубликовано на портале: 30-07-2004
Cambridge, Mass: MIT Press, 1993
Тематический раздел:
Time Series Models is a companion volume to Andrew Harvey's highly successful Econometric Analysis of Time Series. It takes students to another level from the first book, focusing on the estimation, testing, and specification of both univariate and multivariate time series models. The emphasis is on understanding how time series are analyzed and models constructed. Familiarity with calculus, linear algebra, and statistical interference is assumed.
Although Time Series Models pairs well with Harvey's earlier text, it is self-contained. For the second edition, the author has added new sections on nonlinear models, unit roots, structural time series models, intervention analysis, and cointegration. He has addressed new developments, rearranged some material, and changed the emphasis in certain areas.

Preface to the Second Edition

From the Preface to the First Edition



1 Introduction

    1.1 Analysing and Modelling Time Series
    1.2 Outline of the Book

2 Stationary Stochastic Processes and their Properties in the Time Domain

    2.1 Basic Concepts
    2.2 Autoregressive Processes
    2.3 Moving Average Processes
    2.4 Mixed Processes
    2.5 Unobserved Components
    2.6 Prediction and Signal Extraction
    2.7 Properties of the Correlogram and Other Sample Statistics
    2.8 Tests for Randomness and Normality

3 Estimation and Testing of Autoregressive-Moving Average Models

    3.1 Introduction
    3.2 Autoregressive Models
    3.3 Moving Average and Mixed Processes
    3.4 Hypothesis Tests and Confidence Intervals
    3.5 Small Sample Properties
    3.6 Model Selection

4 State Space Models and the Kalman Filter

    4.1 State Space Form
    4.2 Filtering, Smoothing and Prediction
    4.3 Gaussian Models and the Likelihood Function
    4.4 Autoregressive-Moving Average Models
    4.5 Regression and Time-Varying Parameters
      Appendix A Properties of the Multivariate Normal Distribution
      Appendix B Matrix Inversion Lemma

5 Time Series Models

    5.1 Introduction
    5.2 Autoregressive-Integrated-Moving Average Models
    5.3 Structural Time Series Models
    5.4 Autoregressive Models
    5.5 Seasonality
    5.6 Seasonal ARIMA and Structural Models
    5.7 Long Memory and Growth Curves
    5.8 Explanatory Variables
    5.9 Intervention Analysis

6 The Frequency Domain

    6.1 Introduction
    6.2 Fixed Cycles
    6.3 Spectral Representation of a Stochastic Process
    6.4 Properties of Autoregressive-Moving Average Processes
    6.5 Stochastic Cycles
    6.6 Linear Filters
    6.7 Estimation of the Spectrum
    6.8 Maximum Likelihood Estimation of Time Series Models
    6.9 Testing
    6.10 Regression in the Frequency Domain
      Appendix A Trigonometric Identities
      Appendix B Orthogonality Relationships
      Appendix C Fourier Transforms

7 Multivariate Time Series

    7.1 Stationary Series and their Properties in the Time Domain
    7.2 Cross-Spectral Analysis
    7.3 Vector Autoregressive-Moving Average Processes
    7.4 Estimation
    7.5 Multivariate ARIMA Modelling
    7.6 Structural Time Series Models 7.7 Co-integration

8 Non-Linear Models

    8.1 Introduction
    8.2 Conditionally Gaussian Models
    8.3 Autoregressive Conditional Heteroscedasticity
    8.4 Stochastic Variance Models
    8.5 Qualitative Observations and Markov Chains
    8.6 Switching Regimes
      Appendix Law of Iterated Expectations

Answers to Selected Exercises


Subject Index

Author Index

Ключевые слова

См. также:
[Компьютерная программа]
Christopher Sims
Econometrica. 1980.  Vol. 48. No. 1. P. 1-48. 
James Hamilton
Thomas B. Fomby, R. Carter Hill