INSEAD Library

Includes bibliographical references and index

Digitized

An Introduction to State Space Time Series Analysis Contents List of Figures List of Tables 1. Introduction 2. The local level model 2.1. Deterministic level 2.2. Stochastic level 2.3. The local level model and Norwegian fatalities 3. The local linear trend model 3.1. Deterministic level and slope 3.2. Stochastic level and slope 3.3. Stochastic level and deterministic slope 3.4. The local linear trend model and Finnish fatalities 4. The local level model with seasonal 4.1. Deterministic level and seasonal 4.2. Stochastic level and seasonal 4.3. Stochastic level and deterministic seasonal 4.4. The local level and seasonal model and UK inflation 5. The local level model with explanatory variable 5.1. Deterministic level and explanatory variable 5.2. Stochastic level and explanatory variable 6. The local level model with intervention variable 6.1. Deterministic level and intervention variable 6.2. Stochastic level and intervention variable 7. The UK seat belt and inflation models 7.1. Deterministic level and seasonal 7.2. Stochastic level and seasonal 7.3. Stochastic level and deterministic seasonal 7.4. The UK inflation model x xiv 1 9 10 15 18 21 21 23 26 28 32 34 38 42 43 47 48 52 55 56 59 62 63 64 67 70 8. General treatment of univariate state space models 8.1. State space representation of univariate models* 8.2. Incorporating regression effects* 8.3. Confidence intervals 8.4. Filtering and prediction 8.5. Diagnostic tests 8.6. Forecasting 8.7. Missing observations 9. Multivariate time series analysis* 9.1. State space representation of multivariate models 9.2. Multivariate trend model with regression effects 9.3. Common levels and slopes 9.4. An illustration of multivariate state space analysis 10. State space and Box-Jenkins methods for time series analysis 10.1. Stationary processes and related concepts 10.1.1. Stationary process 10.1.2. Random process 10.1.3. Moving average process 10.1.4. Autoregressive process 10.1.5. Autoregressive moving average process 10.2. Non-stationary ARIMA models 10.3. Unobserved components and ARIMA 10.4. State space versus ARIMA approaches 11. State space modelling in practice 11.1. The STAMP program and SsfPack 11.2. State space representation in SsfPack* 11.3. Incorporating regression and intervention effects* 11.4. Estimation of a model in SsfPack* 11.4.1. Likelihood evaluation using SsfLikEx 11.4.2. The score vector 11.4.3. Numerical maximisation of likelihood in ox 11.4.4. The EM algorithm 11.4.5. Some illustrations in ox 11.5. Prediction, filtering, and smoothing* 12. Conclusions 12.1. Further reading APPENDIX A. UK drivers KSI and petrol price 73 73 78 81 84 90 96 103 107 107 108 111 113 122 122 122 123 125 126 128 129 132 133 135 135 136 139 142 144 146 149 150 151 154 157 159 162 APPENDIX B. Road traffic fatalities in Norway and Finland APPENDIX C. UK front and rear seat passengers KSI APPENDIX D. UK price changes Bibliography Index 164 165 167 171 173